FOR PUBLICATION
UNITED STATES COURT OF APPEALS
FOR THE NINTH CIRCUIT
PACIFIC RIVERS COUNCIL,
Plaintiff-Appellant,
v.
UNITED STATES FOREST SERVICE;
MARK REY, in his official capacity
as Under Secretary of Agriculture;
DALE BOSWORTH, in his capacity as
Chief of the United States Forest
Service; JACK BLACKWELL, in his No. 08-17565
official capacity as Regional D.C. No.
Forester, Region 5, United States 2:05-cv-00953-
Forest Service, MCE-GGH
Defendants-Appellees, OPINION
and
CALIFORNIA FORESTRY ASSOCIATION;
AMERICAN FOREST & PAPER
ASSOCIATION; QUINCY LIBRARY
GROUP; PLUMAS COUNTY;
CALIFORNIA SKI INDUSTRY
ASSOCIATION,
Defendants-intervenors-Appellees.
Appeal from the United States District Court
for the Eastern District of California
Morrison C. England, District Judge, Presiding
Argued and Submitted
May 10, 2010—San Francisco, California
Filed February 3, 2012
997
998 PACIFIC RIVERS COUNCIL v. USFS
Before: Stephen Reinhardt, William A. Fletcher and
N. Randy Smith, Circuit Judges.
Opinion by Judge William A. Fletcher;
Dissent by Judge N.R. Smith
PACIFIC RIVERS COUNCIL v. USFS 1001
COUNSEL
Brian Gaffney, LIPPE GAFFNEY WAGNER, San Francisco,
California, Babak Naficy, San Luis Obispo, California, for the
appellant.
Cynthia S. Huber, Joan M. Pepin, U.S. DEPARTMENT OF
JUSTICE, Washington, D.C., Barclay T. Samford, U.S.
DEPARTMENT OF JUSTICE, Denver, Colorado, David
Taylor Shelledy, OFFICE OF THE U.S. ATTORNEY, Sacra-
mento, California, for the appellees.
J. Michael Klise, Thomas Richard Lundquist, CROWELL &
MORING, LLP, Washington, D.C., Steven P. Rice,
CROWLL & MORING, LLP, Irvine, California, Michael B.
Jackson, Quincy, California, for the intervenors-appellees.
OPINION
W. FLETCHER, Circuit Judge:
The national forests of the Sierra Nevada Mountains (“the
Sierras”) are home to a rich array of fauna, including at least
61 species of fish and 35 species of amphibians. The Sierra
Nevada Ecosystem Project, a study commissioned by Con-
gress, concluded in 1996 that their environment has been
severely degraded: “The aquatic/riparian systems are the most
altered and impaired habitats in the Sierra.”
The Sierra Nevada Forest Plan applies to all of the national
forests in the Sierras. In January 2001, the United States For-
est Service (“Forest Service”) issued a Final Environmental
Impact Statement (“2001 EIS”) recommending amendments
to the Forest Plan. The amendments were intended, among
other things, to conserve and repair the aquatic and riparian
ecosystems. In January 2001, under the administration of
1002 PACIFIC RIVERS COUNCIL v. USFS
President Clinton, the Forest Service adopted a modified ver-
sion of the preferred alternative recommended in the 2001
EIS. The parties refer to this as the 2001 Framework.
In November 2001, under the administration of newly
elected President Bush, the Chief of the Forest Service asked
for a review of the 2001 Framework. In January 2004, the
Forest Service issued a Final Supplemental Environmental
Impact Statement (“2004 EIS”) recommending significant
changes to the 2001 Framework. The Forest Service adopted
the preferred alternative in the 2004 EIS. The parties refer to
this as the 2004 Framework.
Plaintiff-Appellant Pacific Rivers Council (“Pacific Riv-
ers”) brought suit in federal district court challenging the
2004 Framework as inconsistent with the National Environ-
mental Protection Act (“NEPA”) and the Administrative Pro-
cedure Act (“APA”). The gravamen of Pacific Rivers’
complaint is that the 2004 EIS does not sufficiently analyze
the environmental consequences of the 2004 Framework for
fish and amphibians. On cross-motions for summary judg-
ment, the district court granted summary judgment to the For-
est Service.
Pacific Rivers timely appealed the grant of summary judg-
ment. For the reasons that follow, we conclude that the Forest
Service’s analysis of fish in the 2004 EIS does not comply
with NEPA. However, we conclude that the Forest Service’s
analysis of amphibians does comply with NEPA. We there-
fore reverse in part, affirm in part, and remand to the district
court.
I. Background
Stretching along a north-south axis for more than 400
miles, the Sierra Nevada Mountains form one of the longest
continuous mountain ranges in the lower 48 states. The Forest
Service manages nearly 11.5 million acres of land under the
PACIFIC RIVERS COUNCIL v. USFS 1003
Sierra Nevada Forest Plan. The Forest Plan is a Land and
Resource Management Plan (“LRMP”) formulated and pro-
mulgated pursuant to the National Forest Management Act
(“NFMA”). See 16 U.S.C. § 1604. NFMA requires the Forest
Service to provide for and to coordinate multiple uses of the
national forests, including “outdoor recreation, range, timber,
watershed, wildlife and fish, and wilderness.” 16 U.S.C.
§ 1604(e)(1). An LRMP adopted pursuant to NFMA guides
all management decisions within the forests subject to that
LRMP. Individual projects are developed according to the
guiding principles and management goals expressed in the
LRMP. See Ohio Forestry Ass’n, Inc. v. Sierra Club, 523 U.S.
726, 729-31 (1998).
The Sierra Nevada Forest Plan applies to the eleven
national forests that run the length of the Sierras from South-
ern California to the California-Oregon border — the
Sequoia, Inyo, Sierra, Stanislaus, Humboldt-Toiyabe,
Eldorado, Tahoe, Plumas, Lassen, and Modoc National For-
ests, and the Lake Tahoe Basin Management Unit. The area
encompassed by the Plan amounts to more than 5% of the
total forest land managed by the Forest Service. See
http://www.fs.fed.us/r5/sierra/about/ (National Forests
encompass 191 million acres). The forests support substantial
economic activity, including logging and grazing, as well as
recreation. The forests comprise dozens of complex eco-
systems. They include iconic natural landmarks such as Mt.
Whitney, Mono Lake, Lake Tahoe, and giant sequoia trees.
As part of its mandate to manage the national forests, the
Forest Service took major steps in the 1990s to improve the
ecological health of the Sierras. In November 1998, the Forest
Service published a Notice of Intent to prepare an Environ-
mental Impact Statement (“EIS”) analyzing a number of pro-
posed changes to the Sierra Nevada Forest Plan. The Forest
Service cited the need to “improve national forest manage-
ment direction for five broad problems: (1) conservation of
old-forest ecosystems, (2) conservation of aquatic, riparian,
1004 PACIFIC RIVERS COUNCIL v. USFS
and meadow ecosystems, (3) increased risk of fire and fuels
buildup, (4) introduction of noxious weeds, and (5) sustaining
hardwood forests.”
In 2000, after nearly a decade of study, the Forest Service
proposed a number of changes to the Forest Plan to ensure
“the ecological sustainability of the entire Sierra Nevada
ecosystem and the communities that depend on it.” The Forest
Service issued a Draft EIS evaluating eight alternatives for
implementing the objectives outlined in the Notice of Intent.
Following public comment, scientific review and consultation
with other agencies, the Forest Service released a Final EIS
in January, 2001.
The 2001 EIS designated the “Modified Alternative 8” as
the preferred alternative. In a Record of Decision issued Janu-
ary 12, 2001, the Forest Service adopted this alternative. This
is the “2001 Framework.”
The Forest Service received over 200 timely administrative
appeals. The Chief of the Forest Service, newly appointed by
the incoming administration, did not respond directly to the
appeals. Rather, he directed the Regional Forester to re-
evaluate the 2001 Framework with respect to three fire-related
issues. First, the Chief directed him “to re-evaluate the deci-
sion for possibilities of more flexibility in aggressive fuels
treatment.” Second, he directed him “to re-evaluate the deci-
sion based on possible new information associated with the
National Fire Plan,” a ten-year strategy developed by Con-
gress, federal agencies, Indian Tribes and western States to
restore fire-adapted ecosystem health. Third, he directed him
to re-evaluate limitations placed by the 2001 Framework on
the Herger-Feinstein Quincy Library Group Forest pilot proj-
ect dealing with fire prevention.
In December 2001, the Regional Forester appointed an
Amendment Review Team. The Regional Forester added non-
fire-related issues to the issues identified by the Chief. In
PACIFIC RIVERS COUNCIL v. USFS 1005
addition to the fire-related issues, he asked the Review Team
to “identify opportunities” in three areas: first, to “reduce the
unintended and adverse impacts [of the 2001 Framework] on
grazing permit holders”; second, to “reduce the unintended
and adverse impacts [of the 2001 Framework] on recreation
users and permit holders”; and, third, to “reduce the unin-
tended and adverse impacts [of the 2001 Framework] on local
communities.”
In June 2003, the Forest Service issued a Draft Supplemen-
tal EIS, based on the work of the Review Team. The Draft
focused on a comparison of two alternatives. “Alternative S1”
was the 2001 Framework. “Alternative S2” was the “preferred
alternative.” Alternative S2 proposed substantially more log-
ging and associated activities than the 2001 Framework. It
also proposed to reduce restrictions on grazing by commercial
and recreational livestock.
The Draft was criticized by the staff of the Forest Service’s
Washington Office for Watershed, Fish, Wildlife, Air and
Rare Plants. The staff wrote a letter complaining that there
was no discussion of the effects of the logging and logging-
related activities on fish:
Aquatic and Riparian: There needs to be a discus-
sion of the effects of the new alternatives on riparian
ecosystems, streams and fisheries. It is not sufficient
to dismiss these effects as within the range of
impacts discussed in the [2001] framework . . . with-
out further analysis, given the activities proposed in
Alternative S2. If the treatments [proposed in Alter-
native S2] will be sufficient to have their intended
effect, there is a high likelihood that there will be
significant and measurable direct, indirect and
cumulative effects on the environment, which need to
be analyzed and disclosed in this document.
1006 PACIFIC RIVERS COUNCIL v. USFS
(Emphasis added.) The letter also raised concerns that the
Draft did not adequately analyze the impact of changed graz-
ing standards on riparian environments, streams and fisheries.
The Forest Service issued the 2004 EIS in January 2004
without adding the discussion of “riparian ecosystems,
streams and fisheries” that the staff letter had said was
needed. The Regional Forester adopted Alternative S2 shortly
afterwards in a Record of Decision. Over 6,000 administrative
appeals were filed objecting to the Record of Decision. The
Forest Service Chief approved the Record Of Decision with-
out change in November 2004. This is the “2004 Frame-
work.”
Both the 2001 and 2004 Frameworks are written in general
terms, rather than addressing specific sites at which the log-
ging and logging-related activities will take place. But there
are substantial differences between the 2001 and 2004 Frame-
works. Relevant to this appeal are changes in authorized log-
ging and logging-related activities, and changes in grazing
standards for commercial and recreational livestock.
The most substantial changes are in logging and logging-
related activities. The 2004 Framework allows the harvesting
of substantially more timber than the 2001 Framework. The
2001 Framework allowed the harvesting of 30 million board
feet of salvage timber per year during the Framework’s first
and second decades. By contrast, the 2004 Framework allows
the harvesting of three times that amount of salvage timber —
90 million board feet per year during its first and second dec-
ades. The 2001 Framework allowed the harvesting of 70 mil-
lion board feet of green timber per year during its first decade
and 20 million board feet per year during its second decade.
By contrast, the 2004 Framework allows the harvesting of 4.7
and 6.6 times that amount of green timber — 329 million
board feet per year during its first decade and 132 million
board feet per year during its second decade. The totals for
salvage timber for the two decades are 600 million board feet
PACIFIC RIVERS COUNCIL v. USFS 1007
under the 2001 Framework, and 1.8 billion board feet under
the 2004 Framework. The totals for green timber for the two
decades are 900 million board feet under the 2001 Frame-
work, and 4.6 billion board feet under the 2004 Framework.
Stated differently, compared to the 2001 Framework, the 2004
Framework allows the harvesting of an additional 4.9 billion
board feet of timber — 1.2 billion board feet of salvage tim-
ber and 3.7 billion board feet of green timber — during its
first two decades.
The 2004 Framework also allows the harvesting of larger
trees than the 2001 Framework. For example, under the 2001
Framework, trees up to 30 inches in breast-height-diameter
could be harvested in the wetter west side of the Sierras, but
only up to 24 inches in the drier east side. Under the 2004
Framework, trees up to 30 inches in breast-height-diameter
can be harvested on both the west and east sides.
The 2004 Framework substantially increases the total acre-
age to be logged. Under the 2004 Framework, about 15%
fewer acres will be subject to prescribed burns than under the
2001 Framework, but about 250% more acres will be logged
“mechanically.” Further, under the 2004 Framework, more
logging will be conducted close to streams than under the
2001 Framework. The 2004 EIS states, with more than the
usual amount of obfuscating bureaucratese:
The spatial location of strategically placed area treat-
ments1 under Alternatives S1 [the 2001 Framework]
and S2 [the 2004 Framework] are the same, but they
are different than previously considered. For exam-
ple, analysis in the [2001 EIS] was based on the
assumption that the area treatments would be placed2
primarily on the upper two-thirds of slopes, thus
1
There is no definitions section in the 2004 EIS. From usage in the EIS,
it is apparent that “treatments” means logging and/or prescribed burns.
2
In standard English, “placed” means “conducted.”
1008 PACIFIC RIVERS COUNCIL v. USFS
minimizing overlap with RCAs3 associated with
perennial, intermittent, and ephemeral streams. How-
ever, this assumption is no longer valid. Conse-
quently, under Alternatives S1 and S2, treatments
are not limited to any geographic position.4 As a
result, more treatments within RCAs are expected.5
Alternative S1 requires that portions of treatment
areas be left in an untreated condition.6 It is likely
that riparian areas would be priorities for retention to
meet this requirement.7 Alternative S2 does not
require retention of untreated areas within treatment
units so that fire behavior and fire effects are effec-
tively reduced within the entire unit.8
3
“RCAs” are Riparian Conservation Areas. See January 2004 Record Of
Decision approving the 2004 EIS, at 114 (“riparian conservation area
(RCA)”).
In its brief to this court, the Forest Service misstates the meaning of the
acronym. It indicates that RCAs are Resource Conservation Areas. See
Response Brief at 33 (“Resource Conservation Areas (‘RCAs’)”). In the
context of this case, the difference between “riparian” and “resource” is
important. “Riparian” is a precise term, meaning something related to the
bank of a river, stream, or other body of water. “Resource” is a general
term, meaning anything from a natural resource such as trees to a financial
resource such as a bank account.
4
This sentence is misleading. “Treatments” (i.e., logging and burning)
under Alternative S1 (the 2001 Framework) are more geographically lim-
ited than “treatments” under Alternative S2 (the 2004 Framework).
5
This sentence translated into standard English: “As a result, more log-
ging and burning close to streams are expected under the 2004 Frame-
work.”
6
This sentence translated into standard English: “The 2001 Framework
requires that certain areas not be logged or burned.”
7
This sentence translated into standard English: “It is likely that under
the 2001 Framework riparian areas would not be logged or burned.”
8
This sentence translated into standard English: “The 2004 Framework
allows logging and burning close to streams in order to eliminate trees
everywhere in a given ‘treatment unit’ as a means of reducing the risk of
fire.”
We remind the Forest Service: “Environmental impact statements shall
be written in plain language . . . so that decisionmakers and the public can
readily understand them. Agencies should employ writers of clear prose
or editors to write, review, or edit statements[.]” 40 C.F.R. § 1502.8.
PACIFIC RIVERS COUNCIL v. USFS 1009
The 2001 Framework limited soil “compaction” in project
areas close to streams to 5% of the area, but the 2004 Frame-
work places no limit on “disturbances” in such areas.
The 2004 Framework allows substantially more construc-
tion of new, and reconstruction of existing, logging roads than
the 2001 Framework. Under the 2001 Framework, 25 miles of
new roads were to be constructed, and 655 miles of existing
roads were to be reconstructed, during the first decade. Under
the 2004 Framework, 115 miles of new roads are to be con-
structed, and 1,520 miles of existing roads reconstructed, dur-
ing the first decade. However, under the 2001 Framework,
950 miles of roads were to be decommissioned, compared
with 1,175 miles of old roads that are to be decommissioned
under the 2004 Framework. The 2004 Framework also allows
an additional 215 miles of temporary roads (43 miles of tem-
porary roads per year for 5 years) and slates an additional
3,200 miles of roads for maintenance (640 miles per year for
five years).
Finally, grazing restrictions under the 2001 Framework are
reduced in the 2004 Framework. Under the 2001 Framework,
commercial livestock (cattle and sheep), as well as recre-
ational livestock (pack and saddle stock used by commercial
outfitters) were to be excluded from meadows known to be
occupied by Yosemite Toads during the toads’ breeding and
rearing seasons, as well as from meadows where surveys to
determine the presence (or absence) of Yosemite Toads had
not yet been performed. The 2004 Framework allows com-
mercial livestock to graze in meadows where surveys to deter-
mine the presence of Yosemite Toads have not yet been
performed. Further, the 2004 Framework eliminates the cate-
gorical exclusion of recreational pack stock and saddle stock
from toad-occupied meadows during the breeding and rearing
season, and allows managers to develop project-based plans
to mitigate effects on the toad.
Other restrictions on grazing have also been reduced. The
2004 Framework divides habitat-protecting restrictions on
1010 PACIFIC RIVERS COUNCIL v. USFS
grazing into several categories based on the adverse impacts
on the grazing permittee: the greater the adverse impact, the
more habitat-protecting effort is required on the part of the
permittee. The 2004 EIS describes the effect of the 2004
Framework on 47 grazing permittees (amounting to 11% of
the “active allotments”). Under the 2001 Framework, there
were no permittees on whom the regulations had no adverse
impact. The regulations had a low adverse impact on 11 per-
mittees, a medium adverse impact on 17, a high adverse
impact on 12, and a very high adverse impact on 7. Under the
2004 Framework, those numbers are, respectively, 14, 7, 10,
9, and 7. That is, a total of 14 grazing permittees who had
been adversely impacted by habitat-protecting regulations
under the 2001 Framework are not adversely impacted at all
under the 2004 Framework. For three of those permittees, the
change effected by the 2004 Framework is to move from a
high adverse impact to no impact at all — that is, to move
from regulations requiring “substantial” habitat-protective
effort by the permittee to regulations requiring no effort what-
soever by the permittee.
The 2004 EIS predicts that the 2004 Framework will
reduce the annual acreage burned by wildfires. Under the
2001 Framework, the estimated annual acreage of wildfires
was 64,000 acres during the first decade, and 63,000 acres
during the fifth decade. Under the 2004 Framework, the esti-
mated annual acreage of wildfires is 60,000 acres during the
first decade, and 49,000 acres during the fifth decade, result-
ing in a total reduction of 18,000 acres over two decades.
Pacific Rivers filed suit in May 2005, alleging that the 2004
Framework was adopted in violation of NEPA and the APA.
On appeal, Pacific Rivers contends that the 2004 EIS fails to
take a “hard look” at the environmental impact of the 2004
Framework on fish and amphibians. We conclude that the
2004 EIS does not comply with NEPA with respect to fish,
but does comply with respect to amphibians.
PACIFIC RIVERS COUNCIL v. USFS 1011
II. Standard of Review
We review de novo questions of Article III justiciability,
including standing. Porter v. Jones, 319 F.3d 483, 489 (9th
Cir. 2003). We also review de novo a district court’s decision
on summary judgment that an agency complied with NEPA.
Or. Natural Desert Ass’n v. Bureau of Land Mgmt., 531 F.3d
1114, 1130 (9th Cir. 2008). Judicial review of an agency’s
compliance with NEPA is governed by the APA, which
requires this court to set aside the agency’s action if it is “ar-
bitrary, capricious, an abuse of discretion, or otherwise not in
accordance with law.” Id. (quoting 5 U.S.C. § 706(2)(A)).
“[W]e will reverse a decision as arbitrary and capricious
only 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.’ ” The Lands Council v. McNair (Lands
Council II), 537 F.3d 981, 987 (9th Cir. 2008) (en banc)
(quoting Earth Island Inst. v. U.S. Forest Serv., 442 F.3d
1147, 1156 (9th Cir. 2006)), overruled on other grounds by
Winter v. Natural Res. Def. Council, 555 U.S. 7 (2008).
“In reviewing the adequacy of an EIS, we employ a rule of
reason to determine whether the EIS contains a reasonably
thorough discussion of the significant aspects of probable
environmental consequences.” Kern v. Bureau of Land
Mgmt., 284 F.3d 1062, 1071 (9th Cir. 2002) (internal quota-
tion marks omitted). “Once an agency has an obligation to
prepare an EIS, the scope of the analysis of environmental
consequences in that EIS must be appropriate to the action in
question. . . . If it is reasonably possible to analyze the envi-
ronmental consequences in an EIS . . . , the agency is required
to perform that analysis.” Id. at 1072.
1012 PACIFIC RIVERS COUNCIL v. USFS
III. Discussion
A. Standing
The Forest Service argues for the first time on appeal that
Pacific Rivers lacks standing under Article III of the Constitu-
tion. Questions of Article III jurisdiction can be raised at any
time. See Ctr. for Biological Diversity v. Kempthorne, 588
F.3d 701, 707 (9th Cir. 2009).
[1] To have standing under Article III, a plaintiff must
establish that
(1) it has suffered an “injury in fact” that is (a) con-
crete and particularized and (b) actual or imminent,
not conjectural or hypothetical; (2) the injury is
fairly traceable to the challenged action of the defen-
dant; and (3) it is likely, as opposed to merely specu-
lative, that the injury will be redressed by a favorable
decision.
Friends of the Earth, Inc. v. Laidlaw Envtl. Servs., 528 U.S.
167, 180-81 (2000). To have standing to seek injunctive relief
under Article III
a plaintiff must show that he is under threat of suf-
fering “injury in fact” that is concrete and particular-
ized; the threat must be actual and imminent, not
conjectural or hypothetical; it must be fairly trace-
able to the challenged action of the defendant; and it
must be likely that a favorable judicial decision will
prevent or redress the injury.
Summers v. Earth Island Inst., 129 S.Ct. 1142, 1149 (2009).
An organization may sue on behalf of its members
when its members would otherwise have standing to
sue in their own right, the interests at stake are ger-
PACIFIC RIVERS COUNCIL v. USFS 1013
mane to the organization’s purpose, and neither the
claim asserted nor the relief requested requires the
participation of individual members in the lawsuit.
Laidlaw, 528 U.S. at 181.
The Forest Service contends that because Pacific Rivers
challenges amendments to a Land and Resource Management
Plan rather than a specific project under the LRMP, it has
failed to allege a threat of a “concrete and particularized”
injury that is “actual or imminent.” The Forest Service also
contends that Pacific Rivers’ members have not specified
which parts of the national forests in the Sierras they use.
The Forest Service relies heavily on the Supreme Court’s
decision in Summers. The plaintiffs in Summers challenged
nationwide regulations promulgated by the Forest Service that
exempted sales of salvage timber of 250 acres or less from
NEPA requirements to prepare an EIS or an Environmental
Assessment (“EA”). Id. at 1147. The plaintiffs initially chal-
lenged a specific sale of salvage timber. After the district
court issued a preliminary injunction, the parties settled the
dispute over that sale. Id. at 1148. On appeal, both before the
Ninth Circuit and before the Supreme Court, the plaintiffs
continued to challenge the validity of the exemption for 250
acres or less, though now there was no specific sale at issue.
They could make only a general statement that they would
visit national forests in the future and might come in contact
with a parcel of 250 acres or less on which a salvage-timber
sale had been conducted without an EIS or an EA. Id. at 1149-
50.
The Supreme Court concluded that there was only a remote
chance, “hardly a likelihood,” that such visits would bring
plaintiffs into contact with land affected by the challenged
regulations. Id. at 1150. The Court noted that the regulation
at issue applied to all national forest land (190 million acres)
and that the size of the affected parcels was small (250 acres
1014 PACIFIC RIVERS COUNCIL v. USFS
or less). Id. “Accepting an intention to visit the National For-
ests as adequate to confer standing to challenge any Govern-
ment action affecting any portion of those forests would be
tantamount to eliminating the requirement of concrete, partic-
ularized injury in fact.” Id.
Summers is substantially different from this case. Pacific
Rivers introduced into evidence in the district court a declara-
tion of its Chairman, Bob Anderson. Anderson declares that
he lives in South Lake Tahoe, that he and his wife own prop-
erty at Mono Lake, and that they “frequently hike and climb
in the Sierra Nevada Range.” Anderson declares further that
Pacific Rivers has over 750 members, some of whom live in
California. He states:
My first Sierra Nevada backpacking trip was to the
Mineral King area in 2000, during which time I also
fished. I plan to continue these activities as long as
the management of Sierra Nevada national forests
does not prevent me from doing so. I have garnered
great personal solace in the knowledge that Sierra
Nevada native species and the watersheds that sup-
port them persist despite over a century’s worth of
impacts from grazing, mining, logging, road build-
ing, dam construction, and related activities. The
same is true for the membership of [Pacific Rivers],
many of whom recreate in, fish throughout, and
derive much satisfaction from the Sierra Nevada.
He writes specifically with respect to members:
[Pacific Rivers] members participate in recreational
activities, such as fishing, hiking, backpacking,
cross-county skiing, nature photography, and river
and lake boating throughout the Sierra Nevada.
The Forest Service challenged Pacific Rivers’s Article III
standing for the first time in this court. If the Forest Service
PACIFIC RIVERS COUNCIL v. USFS 1015
had objected to standing in the district court, Pacific Rivers
could easily have supplemented Anderson’s declaration with
declarations of individual members who use and the enjoy the
Sierras, specifying particular national forests and particular
patterns of use. Given the timing of the Forest Service’s
objection to standing, if we were to hold on the current record
that Pacific Rivers has not sufficiently established threats of
harm to its members who use the Sierras for recreation, we
would remand to the district court to allow further develop-
ment of the record. But we think such additional development
is unnecessary. Anderson has clearly stated that he and a
number of Pacific Rivers’ members have used, and will con-
tinue to use, the national forests in the Sierras in a variety of
places and in a variety of ways.
During the first two decades, the 2004 Framework allows
the harvesting of approximately 4.6 billion board feet of green
timber and approximately 1.8 billion board feet of salvage
timber. This harvesting will take place in every one of the 11
national forests in the Sierras. The smallest amount of green
timber harvesting during the two decades — 35 million board
feet — will take place in the Lake Tahoe Management Basin.
The Lake Tahoe basin is relatively small and is subject to the
most intensive recreational use of the 11 national forests cov-
ered by the 2004 Framework. Anderson lives in the Lake
Tahoe basin. The greatest amount of harvesting — 1.4 billion
board feet — will take place in Plumas National Forest. Har-
vesting in quantities between these two amounts will take
place in each of the other nine national forests covered by the
2004 Framework.
Under the 2004 Framework, much of the timber harvesting
will be in the upper two thirds of slopes, and will therefore
likely be visible from great distances. Significant timber har-
vesting will also take place near streams where recreational
users of the forests are likely to spend much of their time. The
2004 Framework authorizes the construction of 115 miles of
new roads and the reconstruction of 1,520 miles of existing
1016 PACIFIC RIVERS COUNCIL v. USFS
roads during the first decade. Grazing restrictions on commer-
cial and recreational livestock will be reduced throughout the
Sierras.
[2] There is a concrete connection between the interests of
Pacific Rivers’ members in enjoying the forests of the Sierras
and the potential harm caused by the 2004 Framework. See
Lujan v. Defenders of Wildlife, 504 U.S. 555, 562-64 (1992).
There is little doubt that members of Pacific Rivers will come
into contact with affected areas, and that the implementation
of the 2004 Framework will affect their continued use and
enjoyment of the forests. By contrast, the regulation at issue
in Summers affected only small and widely scattered parcels
of land throughout the entire United States, and the plaintiffs
had not shown any realistic likelihood that they would come
into contact with those parcels.
There are two relevant cases in this circuit, both control-
ling. In Salmon River Concerned Citizens v. Robertson, 32
F.3d 1346 (9th Cir. 1994), we held that an environmental
organization had standing to bring a challenge under NEPA
to an LRMP that applied to 6 million acres of national forest
land in the Sierras. Id. at 1349-55. The challenged LRMP
allowed the “use of all methods to treat competing vegetation
. . . [in order] to meet the timber yield objectives,” and dele-
gated the decision to use herbicides to the district foresters. Id.
at 1351.
The Forest Service’s standing argument in Salmon River
was essentially the same as its standing argument here — that
plaintiff lacked standing because it failed “to demonstrate that
the members would be harmed by a specific project using her-
bicides.” Id. at 1352. Members of the organization lived next
to or within the boundaries of the area where herbicides had
previously been banned but would now be permitted, and they
frequently used the area for recreation. Id. at 1353. These
members contended that their health and recreational interests
were adversely affected by the Forest Service’s decision to
PACIFIC RIVERS COUNCIL v. USFS 1017
permit herbicide use. Id. We characterized the members’
injury as the risk “that environmental consequences” of herbi-
cide use “might be overlooked[ ] as a result of deficiencies in
the government’s analysis under environmental statutes.” Id.
at 1355 (internal quotation marks omitted). That risk consti-
tuted a concrete, specific and imminent injury sufficient to
challenge an EIS because “unfettered use of herbicides. . . in
the absence of NEPA compliance will cause harm to visitors’
recreational use and enjoyment, if not to their health.” Id.
We specifically held that the plaintiffs did not have to “wait
to challenge a specific project when their grievance is with an
overall plan.” Id. We explained why:
[I]f the agency action only could be challenged at the
site-specific development stage, the underlying pro-
grammatic authorization would forever escape
review. To the extent that the plan pre-determined
the future, it represents a concrete injury that plain-
tiffs must, at some point, have standing to challenge.
That point is now, or it is never.
Id. (quoting Idaho Conservation League v. Mumma, 956 F.2d
1508, 1516 (9th Cir. 1992)). See also Res. Ltd., Inc. v. Robert-
son, 35 F.3d 1300, 1303 (9th Cir. 1993) (rejecting, in a chal-
lenge to a forest plan, the argument that plaintiffs must “point
to the precise area of the park where their injury will occur”).
[3] Another Ninth Circuit panel has recently addressed a
separate NEPA challenge to the same 2004 Framework at
issue in our case. In Sierra Forest Legacy v. Sherman, 646
F.3d 1161 (9th Cir. 2011), we held that an environmental
organization had standing to challenge the 2004 Framework.
Id. at 1179-80. We noted Ohio Forestry Ass’n v. Sierra Club,
523 U.S. 726 (1998), in which the Supreme Court held that
the Sierra Club’s challenge to an LRMP under the National
Forest Management Act was unripe as a prudential matter, but
did not hold that the Sierra Club lacked Article III standing.
1018 PACIFIC RIVERS COUNCIL v. USFS
Ohio Forestry, 523 U.S. 726. The Court in Ohio Forestry spe-
cifically noted that despite the “considerable legal distance
between the adoption of the Plan and the moment when a tree
is cut, the Plan’s promulgation nonetheless makes logging
more likely in that it is a logging precondition; in its absence
logging could not take place.” Id. at 730. We held in Sierra
Forest Legacy that the harm flowing from a failure to comply
with NEPA in formulating the 2004 Framework was suffi-
cient to confer standing on plaintiff “to bring a facial NEPA
challenge to the 2004 Framework, independent from specific
implementing projects.” 646 F.3d at 1179.
[4] We therefore conclude that Pacific Rivers has Article
III standing to challenge the 2004 Framework under NEPA.
B. NEPA
“The National Environmental Policy Act has ‘twin aims.
First, it places upon [a federal] agency the obligation to con-
sider every significant aspect of the environmental impact of
a proposed action. Second, it ensures that the agency will
inform the public that it has indeed considered environmental
concerns in its decisionmaking process.’ ” Kern, 284 F.3d at
1066 (quoting Baltimore Gas & Elec. Co. v. Natural Res. Def.
Council, Inc., 462 U.S. 87, 97 (1983)) (internal quotations and
citations omitted, alteration in original). NEPA is not substan-
tive. It does not require that agencies adopt the most environ-
mentally friendly course of action. Kern, 284 F.3d at 1066.
Rather, “[t]he sweeping policy goals . . . of NEPA are . . .
realized through a set of ‘action-forcing’ procedures that
require that agencies take a ‘hard look at environmental con-
sequences.’ ” Robertson v. Methow Valley Citizens Council,
490 U.S. 332, 350 (1989) (quoting Kleppe v. Sierra Club, 427
U.S. 390, 410 n.20 (1976)).
Taking a “hard look” at environmental consequences of
major federal actions includes “considering all foreseeable
direct and indirect impacts. Furthermore, a ‘hard look’ should
PACIFIC RIVERS COUNCIL v. USFS 1019
involve a discussion of adverse impacts that does not improp-
erly minimize negative side effects.” N. Alaska Envtl. Ctr. v.
Kempthorne, 457 F.3d 969, 975 (9th Cir. 2006) (internal quo-
tation marks and citations omitted); see also Or. Natural Res.
Council Fund v. Brong, 492 F.3d 1120, 1133 (9th Cir. 2007)
(“[G]eneral statements about possible effects and some risk
do not constitute a hard look absent a justification regarding
why more definitive information could not be provided.”)
(internal quotation marks omitted).
Pacific Rivers alleges that the 2004 EIS does not take a
hard look at environmental consequences of the 2004 Frame-
work on fish and amphibians. For the reasons that follow, we
agree with Pacific Rivers with respect to fish, but disagree
with respect to amphibians.
1. Fish
The 2001 EIS contained a 64-page detailed analysis of
environmental consequences of the 2001 Framework for indi-
vidual species of fish. In stark contrast to the 2001 EIS, the
2004 EIS contains no analysis whatsoever of environmental
consequences of the 2004 Framework for individual species
of fish. The 2004 EIS incorporates by reference the analysis
contained in the 2001 EIS, but contains no analysis of addi-
tional or different environmental consequences of the 2004
Framework even though the new framework authorizes sub-
stantially more environment-altering activities than the old
framework. Of particular importance, the 2004 Framework
allows an additional 4.9 billion board feet of green and sal-
vage timber harvesting during the first two decades, much of
it conducted nearer streams, compared to the 2001 Frame-
work. The 2004 EIS also incorporates by reference two bio-
logical assessments (“BAs”) of the consequences of the 2001
and 2004 Frameworks on listed fish under the Endangered
Species Act. But it neither summarizes the findings of the
BAs nor includes them in an appendix.
1020 PACIFIC RIVERS COUNCIL v. USFS
The Forest Service contends that the 2004 EIS takes a suffi-
ciently hard look at environmental consequences of the 2004
Framework on fish. It makes two arguments. First, it points
out that the 2004 Framework is an amendment to the Sierra
Nevada Forest Plan. The Forest Service argues that because
the Forest Plan is an LRMP, it is not reasonably possible for
the 2004 EIS to provide an analysis of environmental conse-
quences of the 2004 Framework on individual species. Sec-
ond, it argues that the 2004 EIS’s incorporation by reference
of the BAs concerning environmental consequences of the
2001 and 2004 Frameworks on listed fish satisfies the hard
look requirement. We consider these arguments in turn.
a. Level of Required Analysis in the 2004 EIS
Federal law requires preparation of an EIS in conjunction
with the preparation of a programmatic-level plan such as an
LRMP. See 43 C.F.R. § 1601.0-6 (“Approval of a resource
management plan is considered a major Federal action signifi-
cantly affecting the quality of the human environment. The
environmental analysis of alternatives and the proposed plan
shall be accomplished as part of the resource management
planning process . . . .”). The 2004 Framework is not, in itself,
an LRMP; rather, it is an amendment to an LRMP. Some
amendments to LRMPs may be so insignificant that they do
not require preparation of an EIS. But the 2004 Framework is
a fundamental revision of the Sierra Nevada Forest Plan. The
Forest Service does not argue that an EIS is not required. But
the Forest Service does argue that, because of the program-
matic nature of the 2004 Framework, it is not required to per-
form an analysis of environmental consequences for the
individual species of fish.
The required level of analysis in an EIS is different for pro-
grammatic 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:
PACIFIC RIVERS COUNCIL v. USFS 1021
(1) the “programmatic level” at which the
[agency] develops alternative management
scenarios responsive to public concerns,
analyzes the costs, benefits and conse-
quences 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 proj-
ects, 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 pro-
grammatic 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). . . .
Although NEPA requires that the [agency] evalu-
ate the consequences of its action at an early stage in
the project’s planning process, that requirement is
tempered by (1) ‘the statutory command that [a
reviewing court] focus upon a proposal’s parameters
as the agency defines them,’ and (2) ‘the preference
to defer detailed analysis until a concrete develop-
ment proposal crystallized the dimensions of a proj-
ect’s probable environmental consequences.’
[California v. Block, 690 F.2d 753, 761 (9th Cir.
1982)].
Id. at 800-01.
[5] Regardless of whether a programmatic or site-specific
plan is at issue, NEPA requires that an EIS analyze environ-
mental consequences of a proposed plan as soon as it is “rea-
1022 PACIFIC RIVERS COUNCIL v. USFS
sonably possible” to do so. Kern, 284 F.3d at 1072. At issue
in Kern were two things: an EIS for a Resource Management
Plan (“RMP”) for the Coos Bay District in Oregon, and an
Environmental Assessment (“EA”) for a site-specific project
in that district. The RMP in Kern was a programmatic plan,
like the LRMP in the case before us. We wrote:
An agency may not avoid an obligation to analyze
in an EIS environmental consequences that foresee-
ably arise from an RMP merely by saying that the
consequences are unclear or will be analyzed later
when an EA is prepared for a site-specific program
proposed pursuant to an RMP. “[T]he purpose of an
[EIS] is to evaluate the possibilities in light of cur-
rent and contemplated plans and to produce an
informed estimate of the environmental conse-
quences . . . . Drafting an [EIS] necessarily involves
some degree of forecasting.” City of Davis v. Cole-
man, 521 F.2d 661, 676 (9th Cir. 1975) (emphasis
added). If an agency were to defer analysis . . . of
environmental consequences in an RMP, based on a
promise to perform a comparable analysis in connec-
tion with later site-specific projects, no environmen-
tal consequences would ever need to be addressed in
an EIS at the RMP level if comparable consequences
might arise, but on a smaller scale, from a later site-
specific action proposed pursuant to the RMP.
Once an agency has an obligation to prepare an
EIS, the scope of its analysis of environmental con-
sequences 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.
See Save Our Ecosystems v. Clark, 747 F.2d 1240,
1246 n.9 (9th Cir. 1984) (“Reasonable forecasting
and speculation is . . . implicit in NEPA, and we
PACIFIC RIVERS COUNCIL v. USFS 1023
must reject any attempt by agencies to shirk their
responsibilities under NEPA by labeling any and all
discussion of future environmental effects as ‘crystal
ball inquiry,’ ” quoting Scientists’ Inst. for Pub.
Info., Inc. v. Atomic Energy Comm’n, 481 F.2d 1079,
1092 (D.C. Cir. 1973)). If it is reasonably possible to
analyze the environmental consequences in an EIS
for an RMP, the agency is required to perform that
analysis. The EIS analysis may be more general than
a subsequent EA analysis, and it may turn out that a
particular environmental consequence must be ana-
lyzed in both the EIS and the EA. But an earlier EIS
analysis will not have been wasted effort, for it will
guide the EA analysis and, to the extent appropriate,
permit “tiering” by the EA to the EIS in order to
avoid wasteful duplication.
Id. at 1072 (emphasis added). See also 40 C.F.R. § 1501.2
(“Agencies shall integrate the NEPA process with other plan-
ning at the earliest possible time to insure that planning and
decisions reflect environmental values, to avoid delays later in
the process, and to head off potential conflicts.”); New Mexico
ex rel. Richardson v. Bur. of Land Mgmt., 565 F.3d 683, 707-
08, 716 (10th Cir. 2009) (relying on Kern to find NEPA viola-
tion with respect to programmatic EIS).
Our dissenting colleague contends that we overruled Kern
with respect to programmatic-level plans in our en banc deci-
sion in Lands Council II, 537 F.3d 981. We do not believe
that Lands Council II overruled the “reasonably possible”
requirement of Kern. At issue in Lands Council II was an EIS
for a site-specific project. In our en banc opinion, we specifi-
cally overruled Ecology Center, Inc. v. Austin, 430 F.3d 1057
(9th Cir. 2005), cert. denied sub nom. Mineral County v.
Ecology Ctr., Inc., 549 U.S. 1111 (2007). Our holding in
Lands Council II was that the analysis in the site-specific EIS
at issue was sufficiently supported by studies and on-the-
ground analysis. Our opinion nowhere mentioned Kern,
1024 PACIFIC RIVERS COUNCIL v. USFS
nowhere mentioned a programmatic EIS, and nowhere sug-
gested that environmental consequences need not be analyzed
in a programmatic EIS if it is “reasonably possible” to per-
form that analysis.
Nor does the Forest Service believe that Lands Council II
overruled the “reasonably possible” requirement of Kern. The
Forest Service nowhere contends that we wrongly decided
Kern, or that Lands Council II overruled Kern’s “reasonably
possible” requirement. The Forest Service recognizes in its
brief that Kern requires it to perform reasonably possible
analyses of environmental consequences in a programmatic
EIS. See Appellee’s Br. at 25 (“Pacific Rivers correctly notes
that this Court has held [in Kern] that a programmatic EIS
should analyze environmental consequences where ‘reason-
ably possible.’ ”). The Forest Service argues under Lands
Council II that a court owes deference to its determination of
what is reasonably possible because, in its view, “[w]hat sci-
entific analysis is ‘reasonably possible’ at the programmatic
stage is a methodological question within the expertise of the
agency.” Id. But the Forest Service nowhere argues that it
need not comply with Kern.
The 2004 EIS at issue in this case recommends extensive
changes to the 2001 Framework and even more extensive
changes to the underlying Sierra Nevada Forest Plan. We
have described the principal changes above. Briefly, they
include harvesting 4.9 billion more board feet of timber than
under the 2001 Framework (6.4 billion more than under the
Forest Plan). They include logging and burning near streams
that would not have been permitted under the 2001 Frame-
work. They include the construction of 90 more miles of new
roads than under the 2001 Framework (115 more miles than
under the Forest Plan), and reconstruction of 855 more miles
of existing roads than under the 2001 Framework (1,520 more
miles than under the Forest Plan). And they include reduction
of restrictions on grazing by commercial and recreational
stock.
PACIFIC RIVERS COUNCIL v. USFS 1025
[6] The 2004 EIS contains no analysis of environmental
consequences of these changes on individual fish species in
the Sierra. The 2004 EIS promises, in Section 4.2.3 (“Aquatic,
Riparian, and Meadow Ecosystems”), that it will provide such
an analysis. The EIS states, “Effects of the alternatives on
species dependant on aquatic, riparian, and meadow habitats
are explained elsewhere in this [EIS] (Section 4.3.2).”
(Emphasis added). But that promise is not fulfilled. Section
4.3.2 contains 67 pages of analysis of the environmental con-
sequences of the framework for a number of individual spe-
cies of mammals, birds and amphibians who are dependant on
riparian and meadow habitats in the Sierras. But nowhere in
that section (or anywhere else in the 2004 EIS) is there any
analysis of individual species of fish. The explicit promise to
analyze effects “on species dependent on aquatic . . . habitats”
in Section 4.3.2, and the absence of any such analysis in that
section (or anywhere else), is puzzling. It is possible that the
absence of the promised analysis is nothing more than a sim-
ple mistake. But if a mistake, it was a mistake that was specif-
ically brought to the attention of the Forest Service in the
letter written by its Washington staff. As described above,
that letter stated, “There needs to be a discussion of the effects
of the new alternatives on riparian ecosystems, streams and
fisheries.”
In striking contrast to the 2004 EIS, the 2001 EIS contained
64 pages of detailed analysis of environmental consequences
of the 2001 Framework on individual fish species. The 2001
EIS devoted 28 pages to individualized analyses of nine “fed-
erally threatened and endangered fish species” — the Little
Kern Golden Trout, the Paiute Cutthroat Trout, the Lahontan
Cutthroat Trout, the Modoc Sucker, the Warner Sucker, the
Shortnose and Lost River Suckers, the Central Valley Chi-
nook Salmon, and the Central Valley Steelhead Trout. It then
devoted 21 pages to individualized analyses of 11 “sensitive
fish species” — the Goose Lake Lamprey, the Fall Run Chi-
nook Salmon, the Eagle Lake Rainbow Trout, the Volcano
Creek Golden Trout, the Goose Lake Redband Trout, the
1026 PACIFIC RIVERS COUNCIL v. USFS
Warner Valley Redband Trout, the Goose Lake Sucker, the
Lahontan Lake Tui Chub, the Goose Lake Tui Chub, and the
Hardhead. Finally, it devoted 13 pages to individualized anal-
yses of 14 “moderate and high vulnerability fish species” —
the Kern Brook Lamprey, the Pacific Lamprey, the Kern
River Rainbow Trout, the Owens Sucker, the Mountain
Sucker, the Eagle Lake Tui Chub, the Pit River Tui Chub, the
Sacramento Hitch, the Owens Speckled Dace, the Pit River
Roach, the San Joaquin Roach, and the Rough Sculpin.
The 2001 EIS analyzed the environmental consequences to
fish of each of the eight alternatives identified in the EIS. See,
e.g., 2001 EIS, vol. 3, ch. 3, at 262 (“Timber harvesting may
be conducted in riparian areas, following different guidelines,
under Alternatives 3, 4, 6, 7, and Modified 8. Alternatives 3
and 5 prohibit road building in riparian zones; Alternative 5
further addresses negative effects of roads on streams by
requiring that failed road crossings and culverts be identified
and have priority for rehabilitation.”); see also id. at 63, 122
(same). The 2001 EIS also described the environmental con-
sequences of grazing. See, e.g., id. (“One of the greatest risk
factors, within the control of the Forest Service, to Forest Ser-
vice Sensitive fish species in the western United States has
been the degradation of the aquatic environment, especially
those resulting from long term livestock grazing.”); see also
id. at 63, 122 (same).
The 2001 EIS also analyzed particular environmental risks
for individual species of fish. For example, for both Paiute
and Lahontan Cutthroat Trout, “risk factors” included “the
immediate loss of individual fish. . . specific habitat features
such as undercut banks use[d] for cover, increases in sedimen-
tation leading to changes in spawning bed capacity, and the
loss of riparian vegetation necessary to maintain adequate
temperature regime[s].” For Shortnose and Lost River Suck-
ers, risk factors included “[d]ecreases in water quality result-
ing from timber harvest, dredging activities, removal of
riparian vegetation, and livestock grazing.” For Central Val-
PACIFIC RIVERS COUNCIL v. USFS 1027
ley Steelhead, “habitat destruction” was listed as a “risk fac-
tor.” The 2001 EIS noted that “timber harvest, road building,
agriculture, livestock grazing, and urban development” all
“affect[ ] steelhead habitat.” For Volcano Creek Golden
Trout, risk factors included “increases in sedimentation lead-
ing to changes in spawning bed capacity, and the loss of ripar-
ian vegetation necessary to maintain adequate temperature
regime. The risk factors identified are primarily a result of
historic and current grazing practices.” For Goose Lake Suck-
ers, risk factors included the fact that “many of the streams
have experienced some habitat loss due to the effects of log-
ging, grazing and other factors that can degrade watersheds.”
The adequacy of the 2001 EIS with respect to fish is not at
issue. What is at issue is the adequacy of the 2004 EIS.
Whether or not the analysis in the 2001 EIS was adequate (a
question that is not before us), the 2001 EIS shows that an
analysis of environmental consequences of the 2004 Frame-
work for individual species of fish was “reasonably possible.”
There is no explanation in the 2004 EIS of why it was not rea-
sonably possible to provide any analysis whatsoever of envi-
ronmental consequence for individual species of fish, when an
extensive analysis had been provided in the 2001 EIS. There
is also no explanation in the 2004 EIS of why it was “reason-
ably possible” to provide an extensive analysis of environ-
mental consequences to individual species of mammals, birds,
and amphibians in 2004, but not reasonably possible to pro-
vide any analysis whatsoever of environmental consequences
to individual species of fish in 2004.
An agency has flexibility in deciding when to perform
environmental analyses. But an environmental analysis must
“provide ‘sufficient detail to foster informed decision-
making,’ ” Friends of Yosemite Valley, 348 F.3d at 800 (cita-
tion omitted), and so cannot be unreasonably postponed. In
2002, the Council on Environmental Quality (“CEQ”) estab-
lished a Task Force to review agency practices under NEPA.
The Task Force wrote in its September 2003 report to CEQ,
1028 PACIFIC RIVERS COUNCIL v. USFS
“Reliance on programmatic NEPA documents has resulted in
public and regulatory agency concern that programmatic
NEPA documents often play a ‘shell game’ of when and
where deferred issues will be addressed, undermining agency
credibility and trust.” THE NEPA TASK FORCE, MODERNIZING
NEPA IMPLEMENTATION 39 (2003), available at
http://ceq.hss.doe.gov/ntf/report/frontmats.pdf. An agency’s
compliance with the “reasonably possible” requirement in a
programmatic EIS, resulting in an appropriate level of envi-
ronmental analysis, ensures that a “shell game” or the appear-
ance of such a game is avoided. Judicial review under the
arbitrary and capricious standard of the Administrative Proce-
dure Act, 5 U.S.C. § § 706(2)(A), in turn ensures that an
agency does not improperly evade its responsibility to per-
form an environmental analysis when such an analysis is “rea-
sonably possible.”
In some cases, the appropriate level of environmental anal-
ysis in a programmatic EIS is fairly debatable. In such cases,
our obligation is to defer to the expertise of the agency. But
in this case the Forest Service has largely resolved the debate
for us. In its 2001 EIS, the Forest Service performed an exten-
sive analysis of the likely environmental impact of the 2001
Framework, including 64 pages of detailed analysis of the
likely impact on individual fish species. In stark contrast, the
Forest Service performed no analysis whatsoever in its 2004
EIS of the likely impact of the 2004 Framework on fish. The
Forest Service provided no analysis despite the fact that the
2004 Framework allows much more logging, burning, road
construction, and grazing than the 2001 Framework, and
despite the fact that it had provided a detailed analysis in a
programmatic EIS only three years earlier.
We do not require the Forest Service to provide in the 2004
EIS precisely the same level of analysis as in its 2001 EIS.
We recognize that it may be appropriate to have fewer than
64 pages of detailed analysis of environmental consequences
for individual species of fish in the 2004 EIS. Indeed, if the
PACIFIC RIVERS COUNCIL v. USFS 1029
Forest Service had explained its reasons for entirely omitting
any analysis of the impact of the 2004 Framework on individ-
ual species of fish, it is conceivable that it could have con-
vinced us that there is good reason entirely to postpone such
analysis until it makes a site-specific proposal. But the Forest
Service has provided no explanation. Compare 40 C.F.R.
§ 1502.22 (requiring that an agency “always make clear” if it
lacks information to conduct environmental analysis). The
Forest Service has provided almost the opposite of an expla-
nation, for it promised such an analysis and then failed to pro-
vide it. As we noted above, Section 4.2.3. of the 2004 EIS
promises an analysis of the “[e]ffects of the alternatives on
species dependent on aquatic, riparian, and meadow habitats”
in Section 4.3.2. Section 4.3.2 contains a detailed analysis of
the environmental effects on individual species of mammals,
birds and amphibians. But Section 4.3.2. contains no analysis
whatsoever of individual species of fish, even though fish are
the quintessential “species dependant on aquatic . . . habi-
tat[ ].”
[7] In light of the extensive analysis of the environmental
consequences on individual fish species in the 2001 EIS, and
of the extensive analysis of the environmental consequences
on individual species of mammals, birds, and amphibians in
the 2004 EIS, we conclude, contrary to the Forest Service’s
contention, that it was “reasonably possible” to provide some
analysis of the environmental consequences on individual fish
species in the 2004 EIS. The failure of the 2004 EIS to pro-
vide any such analysis is a failure to comply with the hard
look requirement of NEPA.
b. Incorporation by Reference of the Biological
Assessment
The Forest Service’s fall-back argument is that even if an
analysis of environmental consequences of the 2004 Frame-
work for individual fish species was “reasonably possible,”
the hard-look requirement is satisfied by two Biological
1030 PACIFIC RIVERS COUNCIL v. USFS
Assessments (“BAs”), incorporated by reference in the 2004
EIS. We disagree.
[8] Section 7 of the Endangered Species Act requires a
federal agency to consult with the U.S. Fish and Wildlife Ser-
vice (“USFWS”) if a proposed action by that agency “may
affect” a “listed” species or its critical habitat. 16 U.S.C.
§ 1536(a)(2); 50 C.F.R. § 402.14(a). Pursuant to Section 7,
the Forest Service sent two BAs to the USFWS to initiate the
consultation process. The first BA, sent in December 2000,
indicated that the alternatives considered in the 2001 EIS
“may affect” the Little Kern Golden Trout, California Golden
Trout, Lahontan Cutthroat Trout, Paiute Cutthroat Trout,
Owen’s Tai Chub, Modoc Sucker, Lost River Sucker, Short-
nose Sucker, and Warner Sucker. The second BA, sent in July
2003, indicated that the alternatives considered in the 2004
EIS “may affect” all of the species listed in the 2000 BA
except the California Golden Trout.
The 2004 EIS does not include the texts of the BAs, but it
refers to them twice, once in the text and once in an appendix.
First, Section 4.3.1 discusses “Threatened, Endangered, and
Proposed Species.” With respect to the BAs, it states, in its
entirety:
[T]he biological assessment[s] for the [2001 EIS]
and for the [2004 EIS] contain a more thorough anal-
ysis of effects and was [sic] used in evaluating
effects on each species. They are hereby incorpo-
rated by reference.
The text does not identify the individual species of fish
included in the BAs.
Second, Appendix C of the 2004 EIS is a “Consistency
Review” that compares the 2001 and 2004 Frameworks to
determine whether a supplemental environmental analysis is
needed in the 2004 EIS. With respect to “Endangered, Threat-
PACIFIC RIVERS COUNCIL v. USFS 1031
ened, and Proposed Species” of fish, the Consistency Review
concluded:
Implementing the proposed changes considered in
the [2004 EIS] would not be expected to produce
appreciably different results. Effects on these species
are documented in the Biological Assessment for the
[2004 EIS], July 30, 2003.
The Appendix identifies the species of fish covered by the
2003 BA.
The Forest Service’s argument fails for three independently
sufficient reasons.
[9] First, depending on its nature, material should be in the
text of an EIS, should be in an appendix to the EIS, or should
be incorporated by reference in the EIS. In descending order
of importance: (1) Discussion of significant environmental
impacts must appear in the text of an EIS. 40 C.F.R. § 1502.1.
(2) Material that “substantiates any analysis fundamental to
the [EIS]” may appear in an appendix. Id. § 1502.18. (3)
Material may be incorporated by reference so long as its
omission from the EIS does not “imped[e] agency and public
review.” Id. § 1502.21; see also Forty Most Asked Questions
Concerning CEQ’s National Environmental Policy Act Regu-
lations, 46 Fed. Reg. 18026, 18033-34 (March 17, 1981)
(“FAQs”). If the BAs were intended to serve as the analysis
of the environmental consequences of the 2004 Framework
for fish, the 2004 EIS needed to do more than incorporate
them by reference. They should have been described and ana-
lyzed in the text of the 2004 EIS, and the BAs themselves
should have been included in an appendix.
This is not a mere formality. The purpose of an EIS is to
inform decisionmakers and the general public of the environ-
mental consequences of a proposed federal action. That pur-
pose would be defeated if a critical part of the analysis could
1032 PACIFIC RIVERS COUNCIL v. USFS
be omitted from an EIS and its appendices. The EIS is circu-
lated to the general public. “If at all possible,” the appendices
are also circulated to the public. Id. at 18034 (FAQ 25a). The
material that is incorporated by reference is not circulated to
the public; it need only be “made available.” Id. Material that
is incorporated by reference must be “briefly described” in the
body of the EIS, 40 C.F.R. § 1502.21, but a brief description
cannot fulfill the purpose of the EIS if the substance of what
is incorporated is an important part of the environmental anal-
ysis.
Second, even if they had been fully described and analyzed
in the 2004 EIS, the BAs could not have satisfied the “hard
look” requirement. The BAs functioned as a trigger to the
consultation process required under Section 7 of the Endan-
gered Species Act. They merely enumerated the several spe-
cies of “listed” fish that may have been affected by the
alternatives considered in the 2001 and 2004 EISs. There was
no analysis in either of the BAs of the manner or degree to
which the alternatives may have affected these fish. To the
degree that any analysis was performed, it was performed by
the Fish and Wildlife Service when it prepared Biological
Opinions in response to the BAs. The 2004 EIS makes no ref-
erence, in any form, to either of the Biological Opinions.
Third, even if the BAs could have satisfied the hard look
requirement, they applied to only one group of fish species.
As described above, the 2001 EIS analyzed the environmental
consequences for three groups: (1) “federally threatened and
endangered fish species” (9 species); (2) “sensitive fish spe-
cies” (11 species); and (3) “moderate and high vulnerability
fish species” (14 species). The BAs analyzed only the individ-
ual species in the first group. They said nothing whatsoever
about the individual species in the second and third groups.
2. Amphibians
The 2004 EIS contains an extensive analysis of individual
amphibians. It specifically analyzes six species of amphibian:
PACIFIC RIVERS COUNCIL v. USFS 1033
the California Red-legged Frog, the Foothill Yellow-legged
Frog, the Mountain Yellow-legged Frog, the Northern Leop-
ard Frog, the Cascades Frog and the Yosemite Toad. For each
species, the 2004 EIS identifies changes between the 2001
and the 2004 Frameworks that are likely to affect that species.
The 2004 EIS discusses the impact of livestock grazing, pre-
scribed fire, mechanical fuels treatments and road mainte-
nance.
One of the major differences between the 2001 and 2004
Frameworks is the latter’s emphasis on logging, rather than
prescribed burning, as a means of reducing the risk of wild-
fires. The 2004 EIS describes the impact of the changed
emphasis on the Foothill Yellow-legged Frog. It states that the
2001 Framework posed some risk to the frog because pre-
scribed burning often results in the destruction or dispersal of
coarse woody debris that the frog uses for shelter. By decreas-
ing the amount of prescribed burning, the 2004 Framework
will provide some benefit to the frog. However, the 2004 EIS
also identifies the use of mechanical logging as a risk. For
example, the frogs sometimes seek shelter beneath parked
vehicles. When logging operations begin on any particular
day, the vehicles may crush frogs sheltered beneath the tires.
Similarly, the 2004 EIS considers the impact of changed
grazing standards on the Yosemite Toad. It states that risk
factors to the Yosemite Toad from grazing include
decreased growth rate of tadpoles as a result of
increased bacteria from livestock fecal matter; mor-
tality from being buried by livestock feces; reduced
vegetative hiding cover for metamorphs, juveniles,
and adults, which increases their vulnerability to pre-
dation by snakes and birds; and the collapse of
rodent burrows from livestock hoof punching,
thereby entrapping or burying individuals that use
burrows for hiding cover.
1034 PACIFIC RIVERS COUNCIL v. USFS
The 2004 EIS notes that allowing grazing in meadows that
have not yet been surveyed for Yosemite Toads “may contrib-
ute to localized extirpations.”
The 2004 EIS also discusses a number of mitigation strate-
gies to minimize the environmental consequences of the 2004
Framework. For example, the Forest Service will use “Best
Management Practices” for road construction and mainte-
nance. These practices include designing stream crossings and
replacement stream crossings for a 100-year flood; designing
stream crossings to minimize the diversion of natural stream
flow; and avoiding road construction in wetlands and mead-
ows. The Forest Service will also continuously monitor graz-
ing allotments if site-specific changes around Yosemite Toad
breeding sites are authorized. The 2004 EIS states that such
monitoring will allow the Forest Service to identify and miti-
gate threats to the Yosemite Toad.
[10] Pacific Rivers contends that the Forest Service is
required to provide further analysis of the changes that are
authorized under the 2004 Framework. Pacific Rivers’ con-
tention stems in part from the Forest Service’s decision under
the 2004 Framework to delegate significant decisionmaking
authority to local managers of amphibian habitats. For exam-
ple, in a portion of Section 4.2.3 discussing livestock grazing
on meadows, the 2004 EIS notes that the new framework
makes changes designed to “allow flexibility to design man-
agement practices [to] address local conditions.” However,
we are satisfied that the Forest Service’s analysis was suffi-
cient, at this stage of the process, given that the EIS provides
significant analysis of the environmental effects on amphibi-
ans, and that site-specific projects are not yet at issue.
[11] The Forest Service has repeatedly committed itself to
complying with NEPA for site-specific projects that will be
proposed under the 2004 Framework. For example, in its brief
in this court, it states that “additional NEPA analysis will
occur at the project-level.” See Appellee’s Br. at 22. It states,
PACIFIC RIVERS COUNCIL v. USFS 1035
further, that “because on-the-ground activities such as timber
harvest and road, skid trail, and log landing construction
would not occur prior to a future site-specific decision, the
Forest Service will analyze the site-specific effects of those
activities before allowing them.” Id. at 24. The brief states,
still further, that “[w]hen the Forest Service makes a decision
to authorize or reauthorize grazing on an allotment, it con-
ducts a detailed NEPA analysis, where it can examine the
effects of the particular proposed grazing, considering. . . the
allotment’s location. . . , [and] the timing, scope, and intensity
of proposed grazing.” Id. at 40 (emphasis deleted). The Forest
Service makes similar commitments in the 2004 EIS. For
example, the 2004 EIS states, “Site-specific decisions will be
made on projects in compliance with NEPA. . . following
applicable public involvement and administrative appeal pro-
cedures.” The 2004 EIS states further, “Any site-specific
actions taken to implement direction in the Forest Plan
Amendment would require compliance with NEPA.” We are
confident that when the Forest Service proposes to build,
reconstruct or decommission roads; to conduct a logging or a
prescribed burning operation for fuels management; to allow
pack stock and/or saddle stock into Yosemite Toad-occupied
meadows; to permit commercial livestock to graze near
Yosemite Toad breeding and rearing sites; or to use pesticides
in a riparian conservation area, that it will fully comply with
the NEPA requirements applicable to such site-specific proj-
ects.
Conclusion
In Lands Council II, we wrote that we will hold that an
agency has acted in an arbitrary and capricious manner in pre-
paring an EIS when it has “ ‘entirely failed to consider an
important aspect of the problem.’ ” 537 F.3d at 987. In this
case, the Forest Service “entirely failed to consider” environ-
mental consequences of the 2004 Framework on individual
species of fish. Given the detailed 64-page analysis of the
likely impact on individual species of fish in the 2001 EIS, the
1036 PACIFIC RIVERS COUNCIL v. USFS
complete lack of such analysis of the likely impact on individ-
ual species of fish in the 2004 EIS, and the lack of any expla-
nation in the 2004 EIS why it is not “reasonably possible” to
perform some level of analysis of such impact, we have no
choice but to conclude that the Forest Service failed to take
the requisite “hard look” at environmental consequences of
the 2004 Framework for fish.
[12] We hold that the Forest Service failed to take a hard
look at environmental consequences on fish in the 2004 EIS,
in violation of NEPA. We hold that the Forest Service did
take a hard look at environmental consequences on amphibi-
ans in the 2004 EIS, in compliance with NEPA. We therefore
reverse in part and affirm in part, and remand to the district
court.
REVERSED in part, AFFIRMED in part, and
REMANDED. Costs to Plaintiff-Appellant.
N. R. SMITH, Circuit Judge, dissenting:
The majority “conclude[s], contrary to the Forest Service’s
contention, that it was ‘reasonably possible’ to provide some
analysis of the environmental consequences on individual fish
species in the 2004 EIS,” and thus that the agency’s decision
not to provide this analysis “as soon as it [was] ‘reasonably
possible’ to do so” was arbitrary and capricious. Maj. Op.
1021-22, 1029. In doing so, the majority makes two funda-
mental errors: First, it reinvents the arbitrary and capricious
standard of review, transforming it from an appropriately def-
erential standard to one freely allowing courts to substitute
their judgments for that of the agency. In doing so, the major-
ity disregards our circuit’s long-standing precedent holding
that an agency’s timing of analysis required by the National
Environmental Policy Act (NEPA) is not arbitrary and capri-
cious if it is performed before a critical commitment of
PACIFIC RIVERS COUNCIL v. USFS 1037
resources occurs. The majority instead creates an unclear rule
based on “reasonable possibility” that imposes additional pro-
cedures not required by NEPA on the Forest Service. Such a
rule “leave[s] the agencies uncertain as to their procedural
duties under NEPA, . . . invite[s] judicial involvement in the
day-to-day decisionmaking process of the agencies, and . . .
invite[s] litigation.” Kleppe v. Sierra Club, 427 U.S. 390, 406
(1976).
Second, the majority ignores the tiering framework created
by NEPA. Because the majority ignores such framework, it
fails to differentiate between a site-specific environmental
impact statement (“EIS”) and a programmatic EIS that
focuses on high-level policy decisions. Under NEPA regula-
tions on tiering and Ninth Circuit precedent, a programmatic
EIS requires less detailed analysis than a site-specific EIS.
Therefore, agencies are allowed to defer in-depth analysis
until site-specific projects have been identified. Furthermore,
agencies are given wide latitude in the tiering methodology
they choose to implement, so long as the programmatic EIS
allows for informed decision-making. As a result, courts owe
a high level of deference to the methodological choices of the
agency.
Because the majority’s opinion amounts to an inappropriate
and substantial shift in our NEPA jurisprudence, I must dis-
sent.
I. STANDARD OF REVIEW
Congress enacted NEPA to require agencies to produce an
EIS whenever they engage in a major action that could signif-
icantly affect the environment. 42 U.S.C. § 4332(2)(C). How-
ever, Congress also enacted the Administrative Procedure Act
(APA), which governs our review of an agency’s actions.
Under the APA, we must employ a highly deferential standard
of review when reviewing the Forest Service’s actions in this
case. 5 U.S.C. § 706(2)(A). Unless the Forest Service’s action
1038 PACIFIC RIVERS COUNCIL v. USFS
is “arbitrary, capricious, an abuse of discretion, or otherwise
not in accordance with law,” we may not set it aside. Id.
In Lands Council II, a unanimous en banc decision, we
explained that “[r]eview 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 Coun-
cil II), 537 F.3d 981, 987 (9th Cir. 2008) (en banc) (alteration
in original omitted) (quoting Earth Island Inst. v. U.S. Forest
Serv., 442 F.3d 1147, 1156 (9th Cir. 2006), abrogated on
other grounds by Winter v. Natural Res. Def. Council, 555
U.S. 7 (2008)). We also noted that our circuit’s “environmen-
tal jurisprudence ha[d], at times, shifted away from the appro-
priate standard of review,” prior to 2008. Id. at 988.
Although Lands Council II only explicitly overruled Ecol-
ogy Center, Inc. v. Austin, 430 F.3d 1057 (9th Cir. 2005),
explaining that Ecology Center was a case illustrative of this
error, our correction extended beyond that solitary case. We
referred to the shift in our jurisprudence occurring “in recent
years,” which clearly alludes to multiple incorrect decisions.
Lands Council II, 537 F.3d at 988. Our correction also dealt
with the deference owed to agencies under our “appropriate
standard of review” in general, id., rather than just regarding
studies and on-the-ground analysis, as the majority argues,
Maj. Op. 1023-24. We observed that previous decisions com-
mitted “key errors” by imposing on agencies additional “re-
quirement[s] not found in any relevant statute or regulation”
and by showing insufficient deference to agencies and “their
methodological choices.” Lands Council II, 537 F.3d at 991.
Therefore, we renounced this incorrect jurisprudence where
we engaged in “fine-grained” assessments of agency action.
Id. at 993. We instead observed that this was not the proper
role for courts. Id. Rather, “our proper role is simply to ensure
that the Forest Service made no ‘clear error of judgment’ that
would render its action ‘arbitrary and capricious.’ ” Id. (quot-
ing Marsh v. Or. Natural Res. Council, 490 U.S. 360, 378
PACIFIC RIVERS COUNCIL v. USFS 1039
(1989)). The majority relies on cases decided prior to 2008
that suggest a less deferential role for courts. However, Lands
Council II has irrevocably changed the legal landscape by set-
ting forth the high level of deference owed by courts to
agency action.
Accordingly, an agency’s decision can be set aside “only
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. at 987 (internal quotation marks omit-
ted) (emphasis added); see also Sierra Club v. U.S. Envtl.
Prot. Agency, 346 F.3d 955, 961 (9th Cir. 2003).
The majority argues that “the Forest Service ‘entirely failed
to consider’ environmental consequences of the 2004 Frame-
work on individual species of fish.” Maj. Op. 1035. But
“[w]hether an agency has overlooked ‘an important aspect of
the problem,’ . . . turns on what a relevant substantive statute
makes ‘important.’ ” Or. Natural Res. Council v. Thomas, 92
F.3d 792, 798 (9th Cir. 1996) (quoting Motor Vehicle Mfrs.
Ass’n v. State Farm Mut., 463 U.S. 29, 43 (1983)). As dis-
cussed below in Part II, NEPA is the relevant statute. NEPA
does not require site specific analysis be considered at the
programmatic EIS stage. Rather, NEPA encourages the defer-
ral of such analysis until the issues are ripe and analyzing
them will be most meaningful. Thus, the Forest Service can-
not have failed to consider an aspect of the problem required
by NEPA by following NEPA’s tiered analysis structure and
deferring specific analysis.
In addition, though the majority pays lip service to Lands
Council II’s deferential standard of review, the majority relies
on Kern v. U.S. Bureau of Land Management, 284 F.3d 1062
(9th Cir. 2002) to engage in the same type of “fine-grained”
1040 PACIFIC RIVERS COUNCIL v. USFS
analysis that was rebuked in Lands Council II.1 Specifically,
the majority demands that the agency provide whatever analy-
sis the majority determines is “reasonably possible” “as soon
as it can reasonably be done.” Id. at 1072. However, the
majority is unable to provide any support for this rule for at
least two reasons.
First, relying on Kern to require a programmatic EIS to
include reasonably possible site-specific analysis as soon as
reasonably possible stretches the language from Kern far
beyond the facts of the case.2 Kern did deal with a program-
matic EIS. However, the agency actions at issue there were
site-specific timber sales, constituting a critical commitment
1
The majority attempts to argue that the Forest Service recognizes that
Kern is the correct rule. Maj. Op. 1024 (citing Appellee’s Br. at 25). How-
ever, the Forest Service merely admitted that Pacific Rivers was “cor-
rect[ ]” in how it articulated the holding of Kern. In the same paragraph,
the Forest Service argues that the determination of what analysis should
be given in a programmatic EIS is “a methodological question within the
expertise of the agency.” Appellee’s Br. at 25. Furthermore, even if the
Forest Service did make a concession about a question of law, there is “no
reason why we should make what we think would be an erroneous deci-
sion, because the applicable law was not insisted upon by one of the par-
ties.” United States v. Miller, 822 F.2d 828, 832 (9th Cir. 1987) (quoting
Smith Engineering Co. v. Rice, 102 F.2d 492, 499 (9th Cir. 1938)). “The
rule has been repeated in a variety of circumstances. Even if a concession
is made by the government, we are not bound by the government’s ‘erro-
neous view of the law.’ ” Id. (quoting Flamingo Resort, Inc. v. United
States, 664 F.2d 1387, 1391 n. 5 (9th Cir. 1982)).
2
It is also worth noting that the “as soon as it can reasonably be done”
language appears to have been created whole cloth by the court in Kern.
Id. at 1072. This is also true of Kern’s language, with no citation, asserting
that “[i]f it is reasonably possible to analyze the environmental conse-
quences in an EIS. . . , the agency is required to perform that analysis.”
Id. Until now, this language has yet to be quoted by a subsequent Ninth
Circuit appellate case. Indeed, the only case the majority is able to “dig
up” that applies Kern’s rule is from the Tenth Circuit. See Maj. Op. 1023
(citing New Mexico ex rel. Richardson v. Bureau of Land Mgmt., 565 F.3d
683, 707-08, 716 (10th Cir. 2009)). Even in the context of the facts of
Kern, then, this “reasonably possible” rule appears to be a departure from
our established precedent.
PACIFIC RIVERS COUNCIL v. USFS 1041
of resources. Id. at 1069 (“A ‘concrete plan,’ a ‘specific
undertaking,’ and a ‘site-specific program’ incorporating the
Guidelines, such as we anticipated in [a previous case], are
now before us.”). The programmatic EIS in Kern had specifi-
cally deferred analysis of specific actions to future NEPA
analysis. Id. at 1074. Rather than strike down this deferral as
necessarily arbitrary and capricious, the Kern court merely
looked to the subsequent EA to see whether the EA had suffi-
ciently analyzed the site-specific action. Id. (“The second sen-
tence [in the programmatic EIS] is not an analysis, but rather
a promise of a later site-specific analysis to be performed in
connection with specific projects ‘within the range of the
Port-Orford-cedar.’ The revised EA for the Sandy-Remote
Analysis Area is such a site-specific analysis. The adequacy
of that EA has also been challenged by ONRC. We now turn
to that question.”).
Thus, Kern does not support the proposition that a pro-
grammatic EIS must include any site-specific analysis as soon
as reasonably possible if no critical commitments of resources
have occurred. Kern is rather inapposite to such a rule. Thus,
applying the “reasonably possible” rule to a programmatic
EIS that does not contemplate critical commitments of
resources is not only unsupported by Kern’s holding, it also
eviscerates the NEPA tiering framework discussed in Part II.
Second, such a rule, particularly when applied to a pro-
grammatic EIS, constitutes a dramatic departure from this cir-
cuit’s precedent regarding arbitrary and capricious review.
Our long-standing rule has always been that “NEPA requires
a full evaluation of site-specific impacts only when a ‘critical
decision’ has been made to act on site development—i.e.,
when ‘the agency proposes to make an irreversible and irre-
trievable commitment of the availability of resources to [a]
project at a particular site.’ ” Friends of Yosemite Valley v.
Norton, 348 F.3d 789, 801 (9th Cir. 2003) (quoting California
v. Block, 690 F.2d 753, 761 (9th Cir. 1982)). Until that
“threshold” point, we are required to defer to the methodolog-
1042 PACIFIC RIVERS COUNCIL v. USFS
ical choices of the agency regarding the timing of when site-
specific analysis can reasonably be done. Block, 690 F.2d at
761.
The majority is correct that NEPA regulations encourage
agencies to “integrate the NEPA process with other planning
at the earliest possible time.” Maj. Op. 1023 (quoting 40
C.F.R. § 1501.2 ). But “this court has interpreted these regula-
tions as requiring agencies to prepare NEPA documents, such
as . . . an EIS, ‘before any irreversible and irretrievable com-
mitment of resources.’ ” Metcalf v. Daley, 214 F.3d 1135,
1143 (9th Cir. 2000). This rule is derived from the text of
NEPA itself. See Conner v. Burford, 848 F.2d 1441, 1446
n.13 (9th Cir. 1988) (“The ‘irreversible and irretrievable com-
mitment of resources’ criterion is derived from [NEPA],
which requires an EIS to include a statement of ‘any irrevers-
ible and irretrievable commitments of resources which would
be involved in the proposed action should it be implement-
ed.’ ”). This rule has also proved useful, as explained by envi-
ronmental law scholars, because “without inside knowledge,
[courts] really cannot know the status of various initiatives
under consideration . . . .” James Salzman and Barton H.
Thompson, Jr., Environmental Law and Policy 328 (3d ed.
2010). Thus, “to provide a bright line standard” for “challeng-
ing the timing of EIS preparation . . . courts have required that
preparation of an EIS commence ‘before [an] irreversible and
irretrievable commitment of resources.’ ” Id. at 328-29 (quot-
ing Environmental Defense Fund, Inc. v. Andrus, 596 F.2d
848, 852 (9th Cir. 1979)).
Consequently, in multiple cases, we have explained that an
agency’s timing of its analysis becomes arbitrary and capri-
cious only if the NEPA documents are prepared after an irre-
versible and irretrievable commitment of resources has
occurred. See, e.g., Native Ecosystems Council v. Dombeck,
304 F.3d 886, 893 (9th Cir. 2002) (“[T]he issue we must
decide here is whether the Federal Defendants prepared the
EA too late in the decision-making process, i.e., after making
PACIFIC RIVERS COUNCIL v. USFS 1043
an irreversible and irretrievable commitment of resources.”
(quoting Metcalf, 214 F.3d at 1143)). On the other hand, we
have held that an agency is “free to decide not to [provide
NEPA analysis] up until the time it issued its Decision Notice
for the” specific commitment of resources. Id. at 893 (empha-
sis in original). In other words, an agency cannot have entirely
failed to consider an aspect of a problem before a critical
commitment of resources has taken place, because the agency
still has an opportunity up to that point to provide the neces-
sary analysis. Accordingly, whether analysis is “reasonably
possible” and was provided “as soon as it is reasonably possi-
ble” is wholly irrelevant to the inquiry of whether the timing
of the agency’s analysis was arbitrary and capricious.
The majority cites, but ignores, precedent upholding this
critical commitment of resources threshold. See, e.g., Maj.
Op. 1020-21 (citing Friends of Yosemite Valley, 348 F.3d at
800). The majority instead requires its own preferred timing
for NEPA analysis. Essentially, the majority misunderstands
that there is a wide range of permissible agency action
between what courts hope for as ideal agency actions, and
actions that fall below a much lower threshold, becoming
arbitrary and capricious. See F.C.C. v. Fox Television Sta-
tions, Inc., ___ U.S. ___, 129 S. Ct. 1800, 1810 (2009) (under
arbitrary and capricious review, courts “should ‘uphold a
decision of less than ideal clarity if the agency’s path may
reasonably be discerned.’ ” (quoting Bowman Transp., Inc. v.
Arkansas-Best Freight System, Inc., 419 U.S. 281, 286
(1974))); Texas Clinical Labs, Inc. v. Sebelius, 612 F.3d 771,
775 (5th Cir. 2010) (“An agency’s decision need not be ideal
or even, perhaps, correct so long as not arbitrary or capricious
and so long as the agency gave at least minimal consideration
to the relevant facts as contained in the record.” (internal quo-
tation marks omitted)). The majority’s proposed rule would
turn arbitrary and capricious review on its head and allow
courts to keep agencies on a tight leash, directing agencies
based on what courts view as best, as illustrated by the major-
ity’s decision in this case. While there are certainly times
1044 PACIFIC RIVERS COUNCIL v. USFS
when I would disagree with quality or timing of an agency’s
analysis and would enjoy dictating my own agenda, arbitrary
and capricious review simply provides courts with no warrant
to do so.
In the present case, it is undisputed that the Forest service
has not made a critical commitment of resources regarding
any site-specific projects. The 2004 Framework “do[es] not
provide final authorization for any activity,”3 and “subsequent
and full environmental review [of these site-specific projects]
is contemplated,” Friends of Yosemite Valley, 348 F.3d at 801.4
It only establishes the standards and guidelines under which
future projects permitting such actions must occur. Thus, the
Forest Service’s timing of analysis has not reached the bright-
line threshold upheld by our precedent, and the Forest Ser-
vice’s decision to defer more specific analysis regarding fish
cannot be arbitrary and capricious.
II. THE FOREST SERVICE APPROPRIATELY
UTILIZED A TIERED ANALYSIS STRUCTURE
Because it is irrelevant whether the Forest Service provided
a reasonably possible amount of analysis as soon as reason-
ably possible, the appropriate issues to review are actually 1)
whether the agency’s use of a tiered analysis structure was
arbitrary and capricious, and 2) whether the amount of high-
level analysis in the current programmatic EIS was sufficient
to engage in informed decision-making regarding broad poli-
cies affecting all species, including fish.
3
United States Dept. of Agriculture, Forest Service, Record of Decision,
Sierra Nevada Forest Plan Amendment, Final Supplemental Environmen-
tal Impact Statement 24 (January 2004) [hereinafter Record of Decision].
4
See also WildWest Inst. v. Bull, 547 F.3d 1162, 1168 (9th Cir. 2008)
(holding that NEPA analysis need only be performed before there is “any
irreversible and irretrievable commitment of resources,” and thus the For-
est Service’s decision to pre-mark trees did not irretrievably commit the
Forest Service to a specific course of action and was not arbitrary and
capricious).
PACIFIC RIVERS COUNCIL v. USFS 1045
A. The agency’s use of a tiered analysis structure to
defer in-depth analysis until concrete, site-specific
projects were planned was not arbitrary and capri-
cious.
The agency’s methodological decision to utilize a tiered
EIS approach and defer in-depth analysis of site-specific proj-
ects was not only reasonable, but it is also encouraged by the
Council on Environmental Quality’s5 (CEQ) regulations for
implementing NEPA. These regulations explain that
“[a]gencies are encouraged to tier their environmental impact
statements to eliminate repetitive discussions of the same
issues and to focus on the actual issues ripe for decision at
each level of environmental review.” 40 C.F.R. § 1502.20
(citations omitted) (emphasis added). The term “tiering”
refers to “the coverage of general matters in broader environ-
mental impact statements (such as national program or policy
statements)” subsequently followed by “narrower statements
or environmental analyses (such as regional or basinwide pro-
gram statements or ultimately site-specific statements) incor-
porating by reference the general discussions and
concentrating solely on the issues specific to the statement
subsequently prepared.” 40 C.F.R. § 1508.28. These regula-
tions explain that tiering is appropriate when the sequence of
analysis moves from “a program, plan, or policy environmen-
tal impact statement . . . to a site-specific statement or analy-
sis.” § 1508.28(a).
Agencies have a wide range of discretion in determining
how to implement their tiering strategy. In a 2001 memoran-
dum, Frederick Skaer, Director of the Office of NEPA Facili-
tation, explained that “we have deliberately stayed away from
prescriptive guidelines on how to apply tiering so that each
5
The CEQ was established under Title II of NEPA and is charged with
the task of “formulat[ing] and recommend[ing] national policies to pro-
mote the improvement of the quality of the environment.” 42 U.S.C.
§ 4342.
1046 PACIFIC RIVERS COUNCIL v. USFS
tiered process can be custom designed to the specific situa-
tion. You therefore have considerable latitude in the specific
tiering approach you utilize to implement the NEPA policy
mandate of informed decision-making.” Office of NEPA
Facilitation, Memorandum on Tiering of the I-70 Project,
Kansas City, Missouri to St. Louis, June 18, 2001 (citation
omitted); see also Ecology Ctr., Inc. v. Austin, 430 F.3d 1057,
1072 (9th Cir. 2005) (McKeown, J., dissenting) (“The limited
nature of this inquiry underscores the latitude in implementa-
tion and interpretation that Congress intended for its agents.”),
overruled on other grounds by Lands Council II, 537 F.3d at
991.
Because the 2004 Framework is a programmatic EIS, that
focuses on broad policies and general goals and does not
make critical commitments of resources (as discussed in Part
I), the Forest Service’s decision to utilize a tiered approach
and defer more in-depth analysis was clearly a reasonable
choice within the agency’s discretion. Thus, so long as the
programmatic EIS provides sufficient guidelines to foster
informed decision-making (as discussed in Part II.B), nothing
more can be required of the agency at this stage.
The majority acknowledges this NEPA tiering framework.
Maj. Op. 1021-22. Then the majority promptly disregards our
precedent and argues that “[r]egardless of whether a program-
matic 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.” Id. at 1011,
1021-24, 1027-28, 1029, 1036. The majority also argues that
the agency was required to perform an “appropriate level of
environmental analysis” based on what the majority deter-
mines was “reasonably possible.” Id. at 1011, 1023, 1027-29.
The majority observes that the 2001 Framework provided
more analysis of specific aquatic species. Id. at 1019. The
majority also claims that the agency failed to explain why it
provided less analysis of fish in the 2004 Framework. Id. at
1027. As a result, the majority asserts that this proves the
PACIFIC RIVERS COUNCIL v. USFS 1047
agency was able to provide more in-depth analysis than it did.
Id. at 1028-29. Consequently, the majority holds that the
agency’s lesser amount of analysis of fish in the 2004 Frame-
work was arbitrary and capricious. Id. at 1035. The majority’s
arguments suffer from at least four flaws.
First, this is a classic example of courts imposing additional
procedures on agencies that have no basis in statutory or regu-
latory law. Nowhere in the text of NEPA, or its regulations,
is an agency required to provide a similar amount of analysis
in the current EIS as was performed in a previous EIS. Both
the 2001 and the 2004 Frameworks were programmatic envi-
ronmental impact statements. The Forest Service voluntarily
chose to provide more in-depth analysis in the 2001 Frame-
work than was necessary, but nothing in NEPA requires an
agency to provide an equivalent level of analysis for a subse-
quent EIS. As long as the agency’s analysis falls within the
wide zone of reasonability, the agency need not provide the
most ideal analysis in order to avoid having its decision struck
down as arbitrary and capricious. See Dombeck, 304 F.3d at
892 (“We will uphold the Forest Service’s decision not to
[provide NEPA analysis until a later date] unless that decision
was unreasonable.”). While it may irritate the majority that
the Forest Service did not provide as much detailed analysis
in the 2004 EIS as in the 2001 EIS, there is no precedent for
the majority’s decision to strike down the Forest Service’s
decision to defer more in-depth analysis until more concrete
projects have been identified.
NEPA also does not impose a blanket requirement on agen-
cies to provide as much analysis as the majority determines is
reasonably possible “as soon as it can reasonably be done.”
Maj. Op. 1022 (quoting Kern, 284 F.3d at 1072). To the con-
trary, the NEPA regulations about tiering clearly indicate that
delayed analysis is not only allowed, but even preferable in
some instances. 40 C.F.R. § 1502.20; see also Block, 690 F.2d
at 761 (noting that the analysis is more meaningful when a
“concrete development proposal crystallizes the dimensions
1048 PACIFIC RIVERS COUNCIL v. USFS
of a project’s probable environmental consequences”); Ctr.
for Biological Diversity v. U.S. Forest Serv., 349 F.3d 1157,
1166 (9th Cir. 2003) (NEPA’s purpose is “to ensure informed
decision-making to the end that the agency will not act on
incomplete information, only to regret its decision after it is
too late to correct.”).
The majority is correct that NEPA is designed to encourage
agencies to “integrate the NEPA process with other planning
at the earliest possible time.” Maj. Op. 1023 (quoting 40
C.F.R. § 1501.2 ). But, in Friends of Yosemite Valley and
other cases, we have recognized that NEPA’s encouragement
of early analysis is “tempered by (1) ‘the statutory command
that [a reviewing court] focus upon a proposal’s parameters as
the agency defines them,’ and (2) ‘the preference to defer
detailed analysis until a concrete development proposal crys-
tallizes the dimensions of a project’s probable environmental
consequences.’ ” 348 F.3d at 800 (quoting Block, 690 F.2d at
761). The majority ignores this tempering effect. Instead, it
essentially demands as much analysis for fish as the majority
determines is reasonably possible as soon as the agency can
provide it, irrespective of the Forest Service’s methodological
choices and decision to utilize a tiered analysis structure.
Second, the majority’s argument comparing the volume of
analysis between the 2001 and 2004 Frameworks suffers from
the proverbial comparison of apples to oranges. The 2001
Framework contained many more broad-based rules and
clear-cut policies that made for easier identification of issues.
The 2004 Framework by design calls for a flexible approach
based on specific conditions,6 and it leaves critical decisions
6
“In general, the changes proposed in [the 2004 Framework] are
designed to meet the intent of the standards and guidelines in [the previous
Framework], but allow flexibility to design management practices [to]
address local conditions.” United States Dept. of Agriculture, Forest Ser-
vice, Sierra Nevada Forest Plan Amendment, 1 Final Supplemental Envi-
ronmental Impact Statement 214 (January 2004) [hereinafter 1 SEIS]
(emphasis added) (citation omitted).
PACIFIC RIVERS COUNCIL v. USFS 1049
to be made when site specific projects are identified. For
example, under the previous 2001 EIS, the “spacial location
of strategically-placed area treatments” was specifically lim-
ited in geographic location to the “upper two-thirds of
slopes,” whereas the 2004 EIS contains no such geographic
limitations. 1 SEIS at 210. Similarly, the 2001 EIS limited
compaction in riparian conservation areas to “less than 5% of
project activity areas,” whereas the 2004 Framework provides
“[n]o firm numeric standard[s] . . . , thus allowing for site-
specific evaluations.” Id. As a result, it is not surprising that
the 2001 Framework more easily lent itself to more extensive
analysis up front.
Third, the majority’s insistence on requiring the agency to
provide the amount of analysis the majority thinks is appro-
priate as soon as reasonably possible illustrates a misunder-
standing of the tiering framework set forth in the CEQ
regulations. These regulations balance the public’s need to
receive analysis quickly with the public’s competing need to
receive analysis regarding “actual issues ripe for decision.” 40
C.F.R. § 1502.20. To achieve this balance, agencies are given
“wide latitude” in choosing the scope of analysis that will
occur at different stages of the tiered analysis structure. Office
of NEPA Facilitation, Memorandum on Tiering of the I-70
Project, Kansas City, Missouri to St. Louis, June 18, 2001.
The majority correctly observes that the level of analysis
may differ depending on the scope of the agency action. Maj.
Op. 1011, 1023, 1027-29. But then the majority incorrectly
takes it upon itself to determine the scope of the project, based
on the quantity and timing of analysis that the majority deter-
mines is “reasonably possible.” Id. at 1011, 1023. This
approach not only ignores the wide latitude the NEPA regula-
tions accord agencies in determining how to structure their
tiered analysis methodology, it directly contradicts Supreme
Court and Ninth Circuit precedent. See Kleppe, 427 U.S. at
413 (agencies have discretion to “intelligently determine the
scope of environmental analysis and review specific actions
1050 PACIFIC RIVERS COUNCIL v. USFS
[they] may take”); Friends of Yosemite Valley, 348 F.3d at
800 (“[A] reviewing court [must] focus upon a proposal’s
parameters as the agency defines them” (alteration in original
omitted) (quoting Block, 690 F.2d at 761)).
As the majority observes, it is true that the CEQ’s Task
Force has expressed concern that the use of a tiering structure
can result in a “shell game” regarding “when and where
deferred issues will be addressed.” Maj. Op. 1027-28 (citing
The Nepa Task Force, Modernizing Nepa Implementation 39
(2003), available at http://ceq.hss.doe.gov/ntf/report/
finalreport.pdf). But the majority ignores that, in the same
paragraph discussing this potential “shell game,” the Task
Force recommends that the CEQ address the problem by cre-
ating requirements whereby programmatic documents would
“provide a roadmap, explaining where and when deferred
issues raised by the public and/or regulatory agencies will be
addressed.” This potential regulatory solution of requiring a
simple roadmap for programmatic analysis is markedly differ-
ent than the majority’s approach of imposing a novel and
unclear judicial requirement, destroying an agency’s method-
ological flexibility and requiring whatever analysis the major-
ity thinks is “reasonably possible” to be performed “as soon
as it can reasonably be done.”
The majority seems to suggest that the Forest Service inap-
propriately participated in such a “shell game” in this case by
providing a “puzzling” and unfulfilled promise to perform
specific analysis of individual fish species. Maj. Op. 1025.
But, even assuming that the Forest Service was required to
follow through on any promises made in the EIS, the Forest
Service did not break any promises. As the majority acknowl-
edges, the Forest Service never explicitly promised to analyze
individual fish species; it merely explained that the “[e]ffects
of the alternatives on species dependent on aquatic, riparian,
and meadow habitats” would be “explained elsewhere in th[e]
SEIS.” 1 SEIS at 207; see also Maj. Op. 1029. The Forest
Service clearly delivered on this promise. Specifically, as to
PACIFIC RIVERS COUNCIL v. USFS 1051
aquatic habitats, Part II.B highlights the Forest Service’s
extensive analysis regarding how various alternatives would
affect aquatic habitats and the corresponding dependent spe-
cies in general. Moreover, as the majority also notes, the 2004
Framework incorporates by reference two different biological
assessments analyzing the consequences of the 2004 EIS on
individual fish species. Maj. Op. 1029-30. While the analysis
from the biological assessments is likely insufficient for site-
specific NEPA analysis regarding a potential critical commit-
ment of resources affecting fish, it further illustrates that the
Forest Service did not break its promise to provide at least
some analysis of aquatic species in the programmatic EIS.
Fourth, the majority incorrectly asserts that there is “no
explanation” for the Forest Service’s decision to defer more
in-depth analysis of individual fish species. See, e.g., Maj. Op.
1027. However, the Forest Service clearly did explain its rea-
sons for deferring in depth analysis until more site-specific
projects were identified. Specifically, in its Record of Deci-
sion, the Forest Service stated,
Our ability to strategically place fuel treatments for
optimum effectiveness has been compromised by the
set of complicated rules in the [2001 Framework].
The standards and guidelines in that [Framework]
are applied at the stand level, rather than by land
allocations. . . . Some of the rules are so detailed that
they prescribe down to one acre what is allowed, and
require measuring change in canopy to ten percent
increments, which is not consistently practical with
existing measurement tools. This fine-scale
approach limits our ability to make significant prog-
ress. . . . [O]ur ability to strategically place fuels
treatments on the landscape has been compromised
by the complexity of rules [which allows] . . . more
habitat [to be] lost to wildfire. . . . This decision is
intended to reverse that trend.
1052 PACIFIC RIVERS COUNCIL v. USFS
Record of Decision at 8-9; see also Appellee’s Br. at 6. As
a result, the agency explained that the 2004 EIS was being
implemented to “assure the most efficient and appropriate use
of government resources . . . .” Record of Decision at 23-24.
The Forest Service primarily argued not that providing more
analysis would be entirely impossible, but rather that “there
was insufficient information and analytic tools for a meaning-
ful analysis . . . .” Appellee’s Br. at 48 (emphasis added).
Therefore, the majority should have concluded that it was
well within the Forest Service’s discretion to determine that
the benefits of deferring in-depth analysis of aquatic species
to provide more meaningful analysis outweighed any delays
in information.
If the Forest Service commits to a site-specific project in
the future, without engaging in the required level of NEPA
analysis, then Pacific Rivers might have a viable NEPA
claim. Indeed, it is likely that “[t]he deficiencies noted by the”
majority opinion (regarding analysis of fish) “are precisely the
omissions the Forest Service will need to correct in order to
comply fully with NEPA” at a later time. Block, 690 F.2d at
763; see also N. Alaska Envtl. Ctr. v. Lujan, 961 F.2d 886,
891 (9th Cir. 1992) (approving a programmatic EIS that
deferred detailed analysis until an application for a mining
permit was submitted, but noting that “judicial estoppel pre-
cludes the Park Service from later arguing that it has no fur-
ther duty to consider mitigation measures . . .”).
Not only has the Forest Service affirmed many times that
they plan to engage in further detailed analysis when specific
projects are identified,7 but we have a legal duty to assume
7
See, e.g., Record of Decision at 20 (“This [Record of Decision] does
not authorize timber sales or any other specific activity on the Sierra
Nevada national forests. Site-specific decisions will be made on projects
in compliance with NEPA, ESA, and other environmental laws following
applicable public involvement and administrative appeal procedures.”);
United States Dept. of Agriculture, Forest Service, Sierra Nevada Forest
PACIFIC RIVERS COUNCIL v. USFS 1053
that the agency will perform that analysis. In Salmon River
Concerned Citizens v. Robertson, we observed that courts
should “assume that government agencies will . . . comply
with their NEPA obligations in later stages of development.”
32 F.3d 1346, 1358 (9th Cir. 1994) (quoting Conner, 848 F.2d
at 1448).
B. The amount of programmatic, high-level analysis
was sufficient to engage in informed decision-
making regarding broad policies affecting all spe-
cies, including fish.
The majority claims that the Forest Service “entirely failed
to consider an important aspect of the problem” by not pro-
viding in-depth analysis regarding how the 2004 program-
matic Framework would affect specific species of fish. Maj.
Op. 1035 (citing Lands Council II, 537 F.3d at 987). But here,
because the Forest Service chose to utilize a tiered NEPA
analysis structure and implement a programmatic EIS, the rel-
evant scope of “the problem” is whether the Forest Service
“provide[d] ‘sufficient detail to foster informed decisionmak-
ing.’ ” Friends of Yosemite Valley, 348 F.3d at 800 (quoting
Lujan, 961 F.2d at 890-91). As discussed above, the majority
is only able to claim otherwise by ignoring the proper stan-
Plan Amendment, 2 Final Supplemental Environmental Impact Statement,
Response to Public Comments 66, 67 (January 2004) [hereinafter 2 SEIS]
(“Actual locations and miles of roadwork would be determined through
project–level planning and analysis.”); Id. at 124, 125 (“Any site-specific
actions taken to implement direction in the Forest Plan Amendment would
require compliance with NEPA. An environmental analysis would be
completed to assess the potential impacts of proposed activities on water
quality and aquatic and riparian systems. The analysis would also include
an assessment of cumulative watershed effects relative to thresholds of
concern established for watersheds in the project analysis area.”); Appel-
lee’s Br. at 49 (“At the project-level, the Forest Service will consider both
the synergistic effects of actions proposed within a project, where applica-
ble, as well as the cumulative effects of multiple projects conforming to
the Framework, again where applicable.”).
1054 PACIFIC RIVERS COUNCIL v. USFS
dard of review and refusing to defer to the Forest Service’s
discretion in determining the scope of its analysis. See
Kleppe, 427 U.S. at 413 (agencies have discretion to “intelli-
gently determine the scope of environmental analysis and
review specific actions [they] may take”); Friends of Yosemite
Valley, 348 F.3d at 800 (“[A] reviewing court [must] focus
upon a proposal’s parameters as the agency defines them”)
(alteration in original omitted) (quoting Block, 690 F.2d at
761). The scope of analysis in a programmatic EIS can
include considerably less detail than in an EIS analyzing a
site-specific project. See, e.g., Res. Ltd., Inc. v. Robertson, 35
F.3d 1300, 1306 (9th Cir. 1993); Salmon River, 32 F.3d at
1357-58; Block, 690 F.2d at 761.
Thus, under the Forest Service’s tiered-analysis approach,
the 2004 EIS provides sufficient high-level standards to guide
future on-the-ground decisions affecting fish. These standards
generally contemplate the relevant range of potential agency
action and the consequences on various habitats in the Sierra
Nevada. The 2004 Framework “begins by explaining that
cumulative effects were analyzed in detail for the eight alter-
natives considered in the 2001 Framework.” Appellee’s Br. at
50. “It then identifies activities that have occurred” since the
2001 Framework, “including soil and water resource improve-
ments, hazardous fuels reductions, wildfire suppression,” and
road construction. Id.
Specifically regarding aquatic habitats (home to fish spe-
cies), the Framework notes that these are one of the most “de-
graded of all habitats in the Sierra Nevada,” though much of
the original problem was related to “lower elevation dams and
diversions.” 1 SEIS at 3. The EIS observed that “[t]he greatest
effects on the [a]quatic, [r]iparian and [m]eadow [e]cosystems
will generally be from either mechanical fuel treatments or
catastrophic wildfires.” Id. at 12, 96. “Fires can have extraor-
dinary effects on watershed processes and, as a consequence,
significantly influence aquatic organisms and the quality of
aquatic habitats in many ways.” Id. at 208 (citation omitted).
PACIFIC RIVERS COUNCIL v. USFS 1055
These effects include “reductions in riparian shading and
altered streamflows [that] can increase stream temperatures to
extreme levels,” “[f]looding, surface erosion, and mass wast-
ing . . . due to vegetation loss,” and “increases in sedimenta-
tion, debris flows, and wood inputs may occur” as well as
“[c]omplete channel reorganization.” Id.
The Forest Service weighed “tradeoffs between potential
aquatic ecosystem and water quality impacts from fuel man-
agement activities (mechanical treatment and prescribed fire)
and risks associated with high severity wildfires.” Id. (citation
omitted). It recognized that “with respect to aquatic eco-
systems, there are arguments for and against the use of fuels
treatments to reduce the extent and severity of future fires.”
Id. (citation omitted). After providing this analysis, the EIS
determined “alternatives that lower the risk of fire and have
medium levels of treatment pose the least risk to aquatic and
riparian system.” Id. at 12. Therefore, by allowing increased
fuels treatments, the 2004 Framework would reduce the antic-
ipated acres burned by just over 15% from the 2001 Frame-
work. Id. at 98.
The Forest Service recognized that this approach “pose[d]
higher short-term risks to aquatic resources because it pre-
scribes larger amounts of mechanical treatments and greater
treatment intensities.” Id. at 12, 97, 215. But the Forest Ser-
vice concluded that this was mitigated by the expected long-
term benefits to aquatic habitats resulting from reducing wild-
fires. Id. The Forest Service also asserted its intent to reduce
any short-term threats through objectives listed in its “Aquatic
Management Strategy,” best management practices, and goals
related to “landscape-level conditions” and “land allocations”
that would be applied during “project level analysis.” Id. at
12, 97, 207, 210, 215. It was reasonable for the Forest Service
to defer more specific analysis of the proposal’s effect on
aquatic species, because “[p]otential treatment effects on
aquatic, riparian and meadow ecosystems are largely a func-
tion of the amounts, types, intensities, and locations of treat-
1056 PACIFIC RIVERS COUNCIL v. USFS
ments and the standards by which they are implemented.” Id.
at 210.
Although the majority correctly notes that the 2004 Frame-
work anticipates considerably more logging in the forests, the
majority ignores the fact that much of that logging may never
occur. For example, 214 million board feet were offered for
sale on average between FY 2000-2002, but only 118 million
were actually sold—approximately 55%. Id. at 174-75. Simi-
larly, only 58% of the fuel treatments projected under the
2001 Framework were carried out in the first three years of
the Framework. Id.; Appellee’s Br. at 22-23. Therefore, the
Forest Service reasonably concluded that it would be ineffi-
cient to perform a detailed analysis of the impact of activities
that may never take place, and the 2004 EIS contains suffi-
cient analysis of the probable consequences of increased fuel
management at the programmatic level.
The 2004 Framework identified roads as another “critical
component” of the risk and benefit “tradeoffs” to aquatic spe-
cies, which include fish. 1 SEIS at 209. The EIS explained
that roads are just behind wildfires in their potential effect on
“aquatic ecosystems and water quality in forested environ-
ments.” Id. The EIS cited studies discussing how “roads can
deliver more sediment to streams than any other human dis-
turbance in forested environments.” Id. (citation omitted).
However, the studies also indicated that “surface erosion from
roads can be reduced through improved design, construction,
and maintenance practices,” and “[p]roper road location,
drainage, surfacing, and cut slope and fill slope treatments are
important in limiting effects.” Id. (citation omitted). The For-
est Service explained that the proposed “modest reduction in
overall road miles, and improved road conditions,” subse-
quently adopted in the 2004 Framework, were some of “the
most important aspects of reducing risks to aquatic
resources.” Id. at 215.
The Forest Service determined that, because many details
of actual on-the-ground activities were yet unknown, a more
PACIFIC RIVERS COUNCIL v. USFS 1057
detailed analysis would be appropriately conducted when spe-
cific projects were identified. For example, the EIS explained
that “actual locations and miles of roadwork [will] be deter-
mined through project-level planning and analysis.” 2 SEIS at
66. Changing the location of a proposed road by just a few
hundred feet could make a substantial difference in the impact
it had on riparian areas and on fish. A different location might
have significantly different vegetation, soil type, and topogra-
phy. Changing the location could even place a road in a com-
pletely different drainage basin, potentially impacting entirely
different species of fish. See, e.g., Biological Assessment for
SNFPA SEIS 146, July 30, 2003 (Paiute cutthroat trout found
only in 14.5 miles of streams).
The EIS explained that “road management does not vary
substantially between [the 2001 Framework and the 2004
Framework]. Under both alternatives, the . . . biological
effects of roads, as previously described, would be reduced
across the bioregion . . . .” 1 SEIS at 212. The EIS further
noted that, under the 2004 Framework, there would be a
decrease in the net miles of roads. Id. (under the 2004 Frame-
work, “1175 miles would be decommissioned and 115 miles
of new road would be constructed”). Although the miles of
reconstructed roads would almost double and may have short-
term impacts, reconstructed roads would be expected to “im-
prove water quality and aquatic habitat . . . .” Id.
The 2004 EIS also provided analysis of the effects to water-
sheds from on-the-ground activity that the Forest Service
might permit under the Framework. The Framework
explained that, as a broad-based policy, future projects should
remain protective of wildlife but strive for more effective
reduction of hazardous fuels. See, e.g., Appellee’s Br. at 6, 9,
36, 54. It also identified activities that have occurred since the
2001 Framework, including soil and water resource improve-
ments, hazardous fuels reductions, wildfire suppression, and
road construction. Id. at 50. Based on this information, it ana-
lyzed combined or synergistic effects of the elements of the
1058 PACIFIC RIVERS COUNCIL v. USFS
2004 Framework on aquatic ecosystems and species, explain-
ing that the 2001 and 2004 Frameworks are expected to have
similar effects, because both alternatives are required to meet
soil quality standards. Id. at 47-48.
Similarly, the EIS addressed the impacts of grazing with
sufficient detail to satisfy NEPA on a programmatic level. As
with logging and road construction, the Framework calls for
a flexible approach based on specific conditions, rather than
a full-scale analysis at this stage. The same 2001 standards
will continue to be in effect and “are expected to reduce ero-
sion of meadows and improve aquatic habitat conditions by
facilitating the growth of stabilizing vegetation along
streams.” 1 SEIS at 214. The 2001 and the 2004 Frameworks
primarily differ in that changes to utilization and stubble
heights may be allowed in the 2004 Framework when current
range conditions are “good to excellent” (and after “rigor-
ous[ ] evaluat[ion]”). Id. Monitoring requirements under this
flexible approach will “minimize[ ] differences in effects on
aquatic . . . ecosystems between the [2001 and 2004 Frame-
works].” Id.
Thus, after recognizing the general impact that various pro-
posals could have on the environment and the measures that
could mitigate those effects in the programmatic EIS, the For-
est Service reasonably deferred the detailed analysis of future
site-specific projects. Based on this analysis, the Forest Ser-
vice clearly did not “entirely fail[ ]” to consider an important
aspect of the programmatic analysis required to provide
informed decision-making. The majority may have preferred
more specific analysis about individual fish species, but such
preference is not a justifiable reason under NEPA to disregard
the agency’s analysis as arbitrary and capricious.
III. CONCLUSION
The agency clearly did not “rel[y] on factors Congress did
not intend it to consider” when it utilized the tiered methodol-
PACIFIC RIVERS COUNCIL v. USFS 1059
ogy encouraged by the CEQ regulations for implementing
NEPA. Lands Council II, 537 F.3d at 987. The Forest Service
also did not “entirely fail[ ] to consider an important aspect”
of the high level policies set forth in their programmatic EIS.
Id. Lastly, the agency clearly did not offer an explanation for
their programmatic EIS that is “so implausible” that it cannot
“be ascribed to a difference in view or the product of agency
expertise.” Id. Because we can only overturn an agency’s
action if the agency committed one of these arbitrary and
capricious errors, and because no such error occurred in this
case, I would appropriately defer to the Forest Service’s rea-
sonable decision and affirm.