Hapner v. Tidwell

FOR PUBLICATION
UNITED STATES COURT OF APPEALS
FOR THE NINTH CIRCUIT

SHARON J. HAPNER; NATIVE 
ECOSYSTEMS COUNCIL;
ALLIANCE FOR THE WILD ROCKIES,
Plaintiffs-Appellants,
v.
No. 09-35896
TOM TIDWELL, Regional Forester of
Region One of the United States D.C. No.
Forest Service; UNITED STATES  9:08-cv-00092-
FOREST SERVICE, an agency of the DWM
United States Department of OPINION
Agriculture,
Defendants-Appellees,
JANET G. HARTMAN; RONALD E.
HARTMAN,
Defendant-Intervenors-Appellees.

Appeal from the United States District Court
for the District of Montana
Donald W. Molloy, District Judge, Presiding

Argued and Submitted
February 4, 2010—Seattle, Washington

Filed September 15, 2010

Before: William A. Fletcher and Johnnie B. Rawlinson,
Circuit Judges, and Robert S. Lasnik,* Chief District Judge.

*The Honorable Robert S. Lasnik, United States District Judge for the
Western District of Washington, sitting by designation.

14167
14168 HAPNER v. TIDWELL
Opinion by Judge William A. Fletcher
 HAPNER v. TIDWELL 14171

COUNSEL

Timothy M. Bechtold, BECHTOLD LAW FIRM PLLC, Mis-
soula, Montana, Rebecca Kay Smith, PUBLIC INTEREST
DEFENSE CENTER PC, Missoula, Montana, for the appel-
lants.

Brian C. Toth, U.S. DEPARTMENT OF JUSTICE, Washing-
ton, D.C., Mark Steger Smith, OFFICE OF THE U.S.
ATTORNEY, Billings, Montana, for the appellees.

Ronald Walter Opsahl, MOUNTAIN STATES LEGAL
FOUNDATION, Lakewood, Colorado, for the intervenor-
appellee.

OPINION

W. FLETCHER, Circuit Judge:

The United States Forest Service (the “Service”) proposed
the Smith Creek Project (the “Project”) in the Gallatin
National Forest to reduce the risk of severe wildfire, to reduce
the risk of insect infestation and disease, and to promote habi-
tat diversity. Sharon Hapner, Alliance for Wild Rockies, and
Native Ecosystems Council (collectively “Plaintiffs”) chal-
lenged the Project, contending that it violated the National
Environmental Policy Act (“NEPA”) and the National Forest
Management Act (“NFMA”). After a remand, the district
court granted summary judgment to the Service on all of
Plaintiffs’ claims.

We affirm the district court in almost all respects. We
reverse on only one claim, holding that the Project violates
14172 HAPNER v. TIDWELL
NFMA by failing to comply with the elk-cover requirement
contained in the Gallatin National Forest Plan.

I. Background

The Smith Creek Project is located on the west side of the
Crazy Mountains, approximately 35 miles north of Living-
ston, Montana, in the Gallatin National Forest. The Forest is
adjacent to the north and west boundaries of Yellowstone
National Park.

The Project area is forested with lodgepole pine, Douglas
fir, and to a lesser degree aspen, spruce, and sub-alpine fir. It
provides habitat for a number of wildlife species, including
resident and migratory herds of elk and Yellowstone cutthroat
trout. The area historically has experienced large-scale stand-
replacing wildfires that have been important in creating and
maintaining the habitat. It has been previously logged, result-
ing in road construction, skid trail creation, soil disturbance,
and riparian damage.

The Project is near residential areas. The largest of these is
the Smith Creek subdivision, consisting of about 30 summer
and year-round homes. The Service identified a high fire risk
to local residents and firefighters due to fuel build-up and lim-
ited road access to the subdivision. The Service designed the
Project, in part, to reduce likely fire intensity. Other stated
purposes of the Project are to improve habitat diversity by
maintaining meadow and aspen areas, and to reduce the risk
of insect infestations and tree diseases.

The Project would authorize logging on up to 810 acres and
prescribed burning on an additional 300 acres. The logging is
designed to break up “the vertical and horizontal continuity of
vegetation and fuel conditions” in order to reduce the chance
of “crown fires” (fire that climbs to, and spreads through, the
tops of trees) and to slow the spread of wildfires. In some
areas, the logging would remove conifers near aspen trees in
 HAPNER v. TIDWELL 14173
order to promote the growth of aspen groves. In other areas,
the logging would thin trees so that they are spaced 20 to 50
feet apart. The thinning would leave approximately 300-500
irregularly spaced trees per acre. Current tree densities range
up to 3,000 per acre.

Ground-based equipment would be used to cut both large
and small diameter trees on up to 435 acres of the Project,
while helicopters would be used to help remove timber on 145
acres. The work on the remaining 230 acres would consist of
hand-treatment (removal of ladder fuels, limbing of large
diameter trees, and thinning of small diameter trees). In order
to minimize the Project’s impact on soil quality, ground-based
harvest would occur only during the winter and only where
there is frozen ground or sufficient snowpack. No new roads
would be built, but formerly closed roads would be reopened
and used temporarily for the Project.

Some ecosystem restoration activities are part of, or are
associated with, the Project. Road maintenance in 2008
improved drainage on parts of two roads in the Project area,
thereby improving water quality and habitat for Yellowstone
cutthroat trout. That work was completed and is no longer
considered part of the Project. The Project includes additional
road improvements, dependent on the availability of funds,
that would also improve water quality. In addition, five tons
per acre of coarse woody debris would be distributed over 4.1
miles of old skid roads within the Project in order to rehabili-
tate disturbed soil.

In August 2007, the Service issued an environmental
assessment (“EA”) proposing three alternatives for the Proj-
ect. The Service allowed 30 days of public comment, in which
Plaintiffs participated. In December 2007, the Service issued
a finding of no significant impact and a final decision select-
ing the third alternative. On July 1, 2008, Plaintiffs filed suit
in the United States District Court for the District of Montana
claiming that the Project violates NEPA and NFMA.
14174 HAPNER v. TIDWELL
The district court granted in part and denied in part Plain-
tiffs’ motion for summary judgment on October 30, 2008. The
court granted summary judgment to Plaintiffs on their claim
that the Service violated NFMA by failing to map elk habitat
as required by the Gallatin Forest Plan, but granted summary
judgment to the Service on all other claims. The court issued
an injunction against implementation of the Project and
remanded to the Service to conduct the necessary mapping.

In November 2008, the Service issued an EA containing
elk-mapping information and solicited public comment. On
March 6, 2009, the Service issued a second finding of no sig-
nificant impact and final decision re-approving the Project.

Plaintiffs filed a new complaint on June 5, 2009, challeng-
ing the March 2009 decision and raising the same claims
addressed in the earlier case. Rather than opening a new case,
the district court issued an order re-opening the original case
and consolidating it with the newly filed case. The court
explained that it “retains jurisdiction over an injunction, even
in the absence of an express statement to that effect.” The
court granted the Service’s motion to dismiss all duplicative
claims from the earlier decision, and noted that all claims
raised in the first and second complaint would be available on
appeal. On October 8, 2009, the district court held that the
Service had complied with its remand order, granted summary
judgment to the Service, and dissolved the injunction.

Plaintiffs appealed the district court’s dismissal of their
claims, including those disposed of by the court’s order of
October 30, 2008. The parties agree that Plaintiffs’ appeal is
timely as to all claims. See Alsea Valley Alliance v. Dep’t of
Commerce, 358 F.3d 1181, 1184-86 (9th Cir. 2004).

II. Discussion

We review de novo the district court’s grant of summary
judgment. See United States v. City of Tacoma, 332 F.3d 574,
 HAPNER v. TIDWELL 14175
578 (9th Cir. 2003). We review the Service’s actions for com-
pliance with NEPA and NFMA under the Administrative Pro-
cedure Act (“APA”). See Native Ecosystems Council v. U.S.
Forest Serv., 428 F.3d 1233, 1238 (9th Cir. 2005). Under the
APA, the Service’s action may be set aside if it is “arbitrary,
capricious, an abuse of discretion, or otherwise not in accor-
dance with law.” 5 U.S.C. § 706(2)(A). A decision is arbitrary
and capricious if the Service

relied on factors Congress did not intend it to con-
sider, 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 differ-
ence in view or the product of agency expertise.

Lands Council v. McNair, 537 F.3d 981, 987 (9th Cir. 2008)
(en banc) (citation and internal quotation marks omitted). “Al-
though the ultimate scope [of our review] may be narrow, the
depth must be sufficient for us to be able to comprehend the
agency’s handling of the evidence cited or relied upon.” Nw.
Coal. for Alternatives to Pesticides v. EPA, 544 F.3d 1043,
1052 n.7 (9th Cir. 2008) (citation omitted).

A. Plaintiffs’ Claims Under NEPA

NEPA is a purely procedural statute, intended to protect the
environment by fostering informed agency decision-making.
See California ex. rel. Lockyer v. U.S. Dep’t of Agric., 575
F.3d 999, 1012 (9th Cir. 2009). NEPA “does not mandate par-
ticular results, but simply provides the necessary process to
ensure that federal agencies take a hard look at the environ-
mental consequences of their actions.” High Sierra Hikers
Ass’n v. Blackwell, 390 F.3d 630, 639 (9th Cir. 2004) (inter-
nal quotation marks omitted). NEPA requires agencies to pre-
pare a detailed environmental impact statement (“EIS”) for all
“major Federal actions significantly affecting the quality of
the human environment.” 42 U.S.C. § 4332(C). “As a prelimi-
14176 HAPNER v. TIDWELL
nary step, the agency may prepare an Environmental Assess-
ment (‘EA’) to determine whether the environmental impact
of the proposed action is significant enough to warrant an
EIS.” High Sierra Hikers, 290 F.3d at 639-40 (citing 40
C.F.R. § 1508.9). An EA must include “brief discussions” of
the need for the proposal, of reasonable alternatives, and of
the anticipated environmental impacts. See 40 C.F.R.
§ 1508.9(b).

1. Effect on Wildfires

[1] The Service’s EA cites the reduction of the risk of
wildfires to local residents as a primary purpose of the Proj-
ect. Plaintiffs argue that the Service violated NEPA by failing
to address scientific debate concerning whether forest thin-
ning actually reduces wildfire intensity. A failure in an EA to
“discuss and consider” evidence contrary to the Service’s
position suggests that the Service “did not take the requisite
‘hard look’ at the environmental consequences” of its pro-
posed action. Blue Mountains Biodiversity Project v. Black-
wood, 161 F.3d 1208, 1213 (9th Cir. 1998).

This case is different from those in which courts have iden-
tified significant controversies as to the efficacy of the Ser-
vice’s proposed methods. In Sierra Club v. Eubanks, 335 F.
Supp. 2d 1070, 1074, 1077-78 (E.D. Cal. 2004), the Service
failed to respond to scientific studies showing that its activi-
ties would increase fire risk and, instead, relied on studies that
did not actually support its position. In Sierra Club v. Bos-
worth, 199 F. Supp. 2d 971, 979-80 (N.D. Cal. 2002), the lit-
erature review that accompanied the EIS included a report
that called into question the Service’s methodology, but that
the EIS failed to disclose or analyze.

[2] In this case, the EA acknowledges the limits of the ben-
efits that would be provided by the Project. The EA does not
claim that the Project would eliminate wildfires in the area
altogether, but merely that it would reduce potential fire
 HAPNER v. TIDWELL 14177
severity, in particular crown fires. The EA explains that limit-
ing crown fires would enhance firefighter and public safety by
reducing the average rate of fire spread from 1 to 3 miles per
hour to 0.1 to 0.5 miles per hour. The Service’s risk reduction
calculations are supported by studies conducted in other
regions, as well as by extensive modeling.

[3] We therefore affirm the district court’s grant of sum-
mary judgment to the Service on this claim.

2. Global Warming

[4] Plaintiffs argue that the Service violated NEPA by fail-
ing to discuss global warming in the EA. The Service’s deci-
sion to implement a project is arbitrary and capricious under
NEPA if an EA or EIS “entirely failed to consider an impor-
tant aspect of the problem.” Lands Council, 537 F.3d at 987.
However, the Service is only required to focus on the issues
“that are truly significant to the action in question.” 40 C.F.R.
§ 1500.1(b). Also, “[i]mpacts shall be discussed in proportion
to their significance.” 40 C.F.R. § 1502.2(b).

[5] Plaintiffs point out that global warming has been recog-
nized by courts as an issue of national importance. See Mas-
sachusetts v. EPA, 549 U.S. 497, 521 (2007); Ctr. for
Biological Diversity v. Nat’l Highway Traffic Safety Admin.,
538 F.3d 1172, 1221-24 (9th Cir. 2008). They also point out
that the Deputy Chief for the National Forest System has
issued a guidance document directing the Service to incorpo-
rate climate change analysis into its evaluations of projects.
That guidance document suggests, for example, that a qualita-
tive discussion of climate change would be necessary in an
EA for a proposal to underburn 30,000 acres of ponderosa
pine stands. It states, however, that proposals require no dis-
cussion if they are of a “minor scale [so] that the direct effects
would be meaningless.” The Project involves a relatively
small amount of land and it will thin rather than clear cut
trees. Further, we note that the Service addressed comments
14178 HAPNER v. TIDWELL
regarding climate change in its December 2007 notice of final
decision.

[6] We therefore conclude that the EA adequately consid-
ered the Project’s impact on global warming in proportion to
its significance.

3. Soil Quality

[7] The EA acknowledges that the Project would cause
detrimental soil disturbance, particularly through ground-
based logging. Plaintiffs argue that the Service violated
NEPA because it failed to analyze soil disturbance mitigation
measures sufficiently in the EA, and because it should have
prepared an EIS evaluating cumulative soil disturbance. We
reject both arguments.1

[8] The Project seeks to minimize soil disturbance by cut-
ting most of the acreage during the winter such that all
ground-based harvest activities using tractors or skidders
would be done over snow or frozen ground. The Project
would use helicopters on much of the remaining acreage. In
the areas subject to ground-based logging, the Project would
implement soil restoration, placing five tons of woody debris
per acre on old skid roads. The Service provided analytical
support for its claim that placing woody debris on old skid
trails would reduce or eliminate detrimental soil conditions.
The Service’s EA, relying on an expert study, explains that
the method would be “beneficial for restoration of soil pro-
ductivity,” and that it would prevent further soil disturbance
from unauthorized ATV use and from weed infestation. It fur-
ther explains:

This restoration proposal would qualitatively reduce
the detrimental effects of previous harvest. This
1
Plaintiffs also argue that the Project violates NFMA due to soil distur-
bance. We discuss and reject that argument below. See infra Part II.B.1.
 HAPNER v. TIDWELL 14179
method of restoration has been shown to be effective
in Region 1.

The EA’s explanation is not the “perfunctory description of
mitigating measures” provided in Neighbors of Cuddy Moun-
tain v. U.S. Forest Service, 137 F.3d 1372, 1380 (9th Cir.
1998). The Service’s analysis in the EA is consistent with
other agency analysis, unlike in League of Wilderness
Defenders v. Forsgren, 309 F.3d 1181, 1192 (9th Cir. 2002).
Further, the EA relies on applicable Best Management Prac-
tices (“BMPs”), including winter harvesting and placement of
woody debris, that are appropriate for this project. See Envtl.
Prot. Info. Ctr. v. U.S. Forest Serv., 451 F.3d 1005, 1015-16
(9th Cir. 2006) (approving of reliance on specifically identi-
fied BMPs); Cf. Blue Mountains Biodiversity Project, 161
F.3d at 1214 (BMPs not appropriate for specific circum-
stances of project).

[9] We therefore conclude that the EA sufficiently
explained that its mitigation measures would minimize, and
compensate for, any soil disturbance from the Project.

B. Plaintiffs’ Claims under NFMA

[10] NFMA imposes procedural and substantive require-
ments on the Service’s management of national forests. Pro-
cedurally, NFMA requires the Service to develop and
maintain forest resource management plans. See Ecology Ctr.
v. Castaneda, 574 F.3d 652, 656 (9th Cir. 2009) (citing 16
U.S.C. § 1604(a)). After a forest plan is adopted, all subse-
quent agency actions must comply with that plan. Id. The Gal-
latin National Forest Plan (“Gallatin Plan” or “Plan”) is the
relevant forest plan in this case. Substantively, NFMA
requires that the forest plans adopted by the Service provide
certain protections, such as protection of forest habitat and
diversity of wildlife. See 16 U.S.C. § 1604(g)(3).
14180 HAPNER v. TIDWELL
1. Soil Quality

[11] Plaintiffs argue that the Project violates NFMA’s pro-
hibition on timber harvesting that irreversibly damages soil.
Under NFMA, the Service may harvest timber from national
forests only where “soil, slope, or other watershed conditions
will not be irreversibly damaged.” 16 U.S.C.
§ 1604(g)(3)(E)(i). Because the Gallatin Plan does not itself
provide specific standards to implement this requirement, the
Service applies soil standards developed for the Northern
Region as a whole. The Northern Region Soil Quality Stan-
dard prohibits logging that results in more than 15% detri-
mental soil disturbance in affected areas. In areas where more
than 15% detrimental soil conditions already exist from prior
activities, the standard allows harvesting only if it does not
result in a net increase in soil disturbance.

Due to previous logging, 408 acres of the Project already
exceed 15% detrimental soil disturbance. On these acres, and
on an additional 28 acres, the Project would employ ground-
based harvesting, resulting in an additional 3.5% soil distur-
bance on average. This represents the equivalent of an addi-
tional 15 acres of disturbance. The Service estimates that
approximately 15 acres of land would be treated by its woody
debris restoration method, thus resulting in no net increase in
soil disturbance in the areas.

Plaintiffs respond that it is merely speculative that the Ser-
vice will follow through on the soil-disturbance mitigation
measures described in the EA. The Service has not yet set
aside funding for all of its proposed mitigation activities, and
the Service has failed to follow through on mitigation in past
projects. However, the woody debris restoration method is
one of the least expensive proposed mitigation measures,
costing only $12,300 over a five-year period. Most of the
unfunded mitigation methods in the Project are road treat-
ments that are unrelated to soil disturbance. Lack of funding
 HAPNER v. TIDWELL 14181
therefore does not appear to be a serious problem for the soil
restoration plans.

[12] We therefore conclude that it was not arbitrary and
capricious for the Service to conclude that its mitigation mea-
sures will ensure that the Project does not violate the Northern
Region soil standards.

2. Old Growth Indicator Species

[13] The Gallatin Plan requires that designated manage-
ment indicator species (“MIS”) “be monitored to determine
population changes.” The designated MIS for the habitat
impacted by the Project are the northern goshawk and pine
marten. Plaintiffs argue that the Service violated the Gallatin
Plan, and therefore NFMA, by failing to monitor northern
goshawk and pine marten. They also argue that the Service
violated NFMA by not accurately determining the effect of
the Project on those species.2 We reject both arguments.

The EA identifies a number of efforts to monitor the popu-
lations of both the northern goshawk and the pine marten in
the region. Montana conducts annual studies of pine marten
population trends that look at snow tracks of the birds. The
studies have detected an average of 75 pine martens per 100
“transect miles” in southwest Montana over a ten-year period.
A recent Service study on species viability in the Gallatin For-
est concluded that habitat to support the pine marten is abun-
dant in the Project area. The Service has also conducted
habitat surveys and population monitoring for northern gos-
hawks in the Gallatin Forest. The Service relied in its EA on
several independent surveys concluding that goshawk viabil-
ity is not a concern. One 2005 study involved a “systematic
2
Plaintiffs also claim that the Service violated NEPA by failing to take
a “hard look” at the Project’s effects on MIS. Plaintiffs’ NEPA claim
raises no issues not included in its NFMA claim, so we do not analyze the
NEPA claim separately. Cf. Lands Council, 537 F.3d at 994-96.
14182 HAPNER v. TIDWELL
random survey” that “showed that the goshawk is relatively
common and well-distributed in the Northern Region.”
Another study from 2005 concluded that “short-term viability
of the goshawk in the Northern Region is not an issue.”

[14] In evaluating the effect of the Project on MIS popula-
tions, the Service appropriately relied on management of the
species’ habitat as a proxy for management of the species
themselves. In Lands Council, 537 F.3d at 996, we approved
“of the Forest Service’s use of the amount of suitable habitat
for a particular species as a proxy for the viability of that spe-
cies.” We concluded that such a method was appropriate even
where the Service is taking actions that “will disturb some
suitable habitat.” Id. at 999.

[15] We review the Service’s application of the proxy
approach under the arbitrary and capricious standard. Id. at
997-98. The Service “must both describe the quantity and
quality of habitat that is necessary to sustain the viability of
the species in question and explain its methodology for mea-
suring this habitat.” Id. at 998. We conclude that the Service’s
reliance on habitat as proxy in this case was not arbitrary and
capricious. In the EA, the Service adequately demonstrated a
“knowledge of what quality and quantity of habitat is neces-
sary to support the species,” and its “method for measuring
the existing amount of habitat [was] reasonably reliable and
accurate.” Lands Council, 537 F.3d at 998-99 (quotation
omitted). The EA describes in detail the habitat necessary for
the viability of both the pine marten and the goshawk. It ade-
quately describes existing habitat in the Gallatin Forest and
the Project area, as well as the Project’s limited effect on the
habitat’s ability to sustain MIS. Further, the EA provides for
substantial mitigation measures to ensure that a nest is not dis-
turbed in the unlikely event that one is located during imple-
mentation of the Project. The Service reasonably concluded
that sufficient habitat currently exists for both species and that
the Project would have “[n]o direct effects on the pine mar-
ten,” and “little, if any, direct affect [sic] on goshawks.”
 HAPNER v. TIDWELL 14183
Plaintiffs nonetheless argue that the Service’s methodology
for measuring existing habitat was flawed because the Service
failed to adequately field-verify old growth stands. The Ser-
vice conducted extensive examinations of stands in the field
in the 1980s, as well as additional subsequent field verifica-
tion. The Service is not required to use on-the-ground verifi-
cation so long as alternative methodologies are reliable. “To
always require a particular type of proof that a project would
maintain a species’ population in a specific area would inhibit
the Forest Service from conducting projects in the National
Forests.” Lands Council, 537 F.3d at 997. The Service used
multiple databases containing old growth stand information
and multiple methods to verify its old growth information in
the Project area. This analysis led the Service to predict a
range of between 21.7% and 29.3% old growth in Gallatin
Forest, with a 90% confidence interval, which places the old
growth area well above the 10% minimum required by the
Gallatin Plan.

[16] We therefore conclude that the Service’s conclusion
that the Project would comply with the Gallatin Plan’s moni-
toring requirements is not arbitrary and capricious.

3. Yellowstone Cutthroat Trout

[17] The Gallatin Plan requires that “[h]abitat that is essen-
tial for [Yellowstone cutthroat trout] be managed to maintain
the[ ] species.” As the Service recognized in the EA, streams
in the Project area are “extremely important to conservation
and recovery of the species.” Yellowstone cutthroat trout in
the area are at risk from non-native fish and from “habitat
degradation due to roads and timber harvest.” In particular,
the Service noted that “any increase in sediment yield from
this proposal would perpetuate degraded spawning habitat
conditions.” Plaintiffs argue that the Project would violate the
Gallatin Plan by increasing sediment levels in the streams due
to ground disturbance from harvesting and increased road use.
We disagree.
14184 HAPNER v. TIDWELL
[18] The Project, along with its associated road improve-
ments, would reduce rather than increase long-term sediment
levels. Plaintiffs argue that some of the improvements in sedi-
ment levels will result from pre-Project road improvements
that have already been completed, and therefore are no longer
considered part of the Project. But even putting those
improvements aside, the Project would reduce long-term
levels through other road improvements. Plaintiffs argue that
these improvements are not yet fully funded, and therefore not
guaranteed. But Plaintiffs fail to show any likelihood of
increased long-term sediment levels from the Project even if
the road improvements are not completed. We therefore find
that the Service was justified in concluding that the Project
does not present a threat to cutthroat trout through increases
in long-term sediment levels.

[19] The Service has also adequately taken short-term sed-
iment increases into account. We have held agency decisions
to be unlawful where they focused only on long-term effects,
ignoring short-term effects that jeopardize species. See Pac.
Coast Fed’n of Fishermen’s Ass’ns v. U.S. Bureau of Recla-
mation, 426 F.3d 1082, 1092, 1095 (9th Cir. 2005) (the
agency failed to provide analysis showing that “the coho
would receive sufficient protection against jeopardy under the
proposed plan”); Nat’l Wildlife Fed’n v. Nat’l Marine Fish-
eries Serv., 524 F.3d 917, 935 (9th Cir. 2008) (proposal at
issue “would have significant negative impacts on each
affected species’ critical habitat” and the agency analysis
failed to “demonstrate that these impacts would not affect the
fishes’ survival and recovery, in light of their short life-cycles
and current extremely poor habitat conditions”). But the Proj-
ect incorporates a number of mitigation measures intended to
limit short-term increase in sediment levels. All ground-based
harvest activities using tractors or skidders would be done
over snow or frozen ground, limiting soil disturbance that
might eventually reach the streams; the Project would limit
harvest activities within 100 feet of any stream; and the Proj-
 HAPNER v. TIDWELL 14185
ect would leave untouched those trees identified as most
likely to provide stability to soil around streams.

The EA explains that any small short-term increases in sed-
iment levels that nonetheless occur despite the Service’s miti-
gation efforts would not threaten the viability of the
Yellowstone cutthroat trout. Current sediment levels vary by
location. In the most severely affected areas, sediment levels
are predicted to be 26% above natural levels over the short
term. According to Service studies, such levels permit a
stream to provide for a population of cutthroat trout 90% the
size of its inherent capability.

[20] We therefore conclude that the Service was not arbi-
trary and capricious in concluding that the Project would not
threaten Yellowstone cutthroat trout through either long-term
or short-term increases in sediment levels.

4. Road Density

Prior to 2006, the Gallatin Plan limited road density in the
forest in order to protect elk habitat. The Plan ensured at least
70% elk effective cover, limiting road density to 0.75
miles/square mile. It did so based on the recommendations of
scientists who stated that “[r]oads are undoubtedly the most
significant consideration on elk summer range.” The road
density restriction was intended to meet the NFMA require-
ment that the Service “provide for diversity of plant and ani-
mal communities.” 16 U.S.C. § 1604(g)(3)(B); see also 36
C.F.R. § 219.19 (2000) (“Fish and wildlife habitat shall be
managed to maintain viable populations.”).

In 2006, the Gallatin Plan was amended to remove the road
density restriction. The Forest Supervisor concluded that the
restriction “was not scientifically supportable or logical” and
that it caused unjustifiable problems. The Supervisor
explained that the standard compelled more timber harvests in
some areas than was desired, led to disagreements over its
14186 HAPNER v. TIDWELL
application, was ineffective in protecting elk, and was already
violated in many areas. The Plan amendment placed all road
development decisions within the Travel Management Plan
and the Service’s discretion. The current road density in the
Project area is 2.1 miles of road/square mile, which well
exceeds the previous 0.75 limitation.

Plaintiffs argue that the 2006 amendment to the Gallatin
Plan violated NFMA. They claim that the Project therefore
violates NFMA because it would be implemented subject to
an unlawful plan. The Service responds that the 2006 amend-
ment did not violate NFMA because it was justified by new
scientific data and by the circumstances prevailing in the Gal-
latin National Forest.

We need not decide whether the 2006 amendment violated
NFMA because the Project would reduce, not increase, long-
term road density in the area. We consider challenges to the
lawfulness of a forest plan only to the extent that the con-
tested portion of the plan “plays a causal role with respect to
the [Project].” Ohio Forestry Ass’n v. Sierra Club, 523 U.S.
726, 734 (1998); see also Neighbors of Cuddy Mountain v.
Alexander, 303 F.3d 1059, 1067 (9th Cir. 2002). The Project
would use only existing roads, some of which would be per-
manently closed and rehabilitated at the end of the Project.
Because road density in the Project area would not be
increased as a result of the 2006 amendment, we have no
occasion to evaluate its lawfulness.

5. Elk Cover

[21] Plaintiffs’ single meritorious argument on appeal con-
cerns the Gallatin Plan’s elk-cover requirement. The Gallatin
Forest Plan requires that the Service “[m]aintain at least two
thirds of the hiding cover associated with key habitat compo-
nents over time. Subsequent timber sale activity will be
allowed after regeneration provides hiding cover.” Elk are
designated in the Plan as an indicator species of the Gallatin
 HAPNER v. TIDWELL 14187
National Forest, for which two-thirds hiding cover must be
maintained. Plaintiffs argue that the Project violates the Plan
because it would reduce elk cover to under two thirds. We
agree.

[22] In preparing the EA, the Service did not measure elk
cover according to the definition provided in the Gallatin
Plan. The Gallatin Plan defines elk cover as “[v]egetation, pri-
marily trees, capable of hiding 90 percent of an elk seen from
a distance of 200 feet or less.” In the EA, the Service relied
on two separate measurements of elk cover. One calculation
of cover was based on the current prevalence of various tree
classifications in the Project area. The Service measured
70-90% elk cover under this definition but does not explain
what percent cover this translates to under the Plan definition.
The other calculation suggested 62% elk cover under a can-
opy cover definition. According to a table in the Helena
National Forest Plan, 60% elk cover as measured under the
canopy cover definition translates to 42% elk cover as mea-
sured under the Gallatin Plan definition. The Project therefore
violates the Gallatin Plan’s two-thirds elk-cover requirement.
The Service’s failure to measure elk cover as defined by the
Gallatin Plan renders us “unable to determine from the record
that the agency is complying with the forest plan standard.”
Native Ecosystems Council v. U.S. Forest Serv., 418 F.3d 953,
962 (9th Cir. 2005).

The Service argues that the Gallatin Forest Plan does not
require that elk cover exceed 67% at all times, but only that
any Service action retain two thirds of then-existing elk cover.
The Service claims, in other words, that the Plan prohibits
only timber sales that would reduce now-existing elk cover by
more than 33%. The Service bases its argument on the word-
ing of the Plan, which requires the Service to maintain “two
thirds of the hiding cover” as opposed to merely “two thirds
hiding cover.” Because the Project would reduce now-existing
elk cover by less than 33% under any measure, the Service
claims that the Project complies with the Gallatin Plan.
14188 HAPNER v. TIDWELL
“Agencies are entitled to deference to their interpretation of
their own regulations, including Forest Plans.” Native Eco-
systems Council, 418 F.3d at 960. “[W]e have effectively
treated forest plan directives as equivalent to federal regula-
tions adopted under the APA, deferring to the Forest Service’s
interpretation of plan directives that are susceptible to more
than one meaning unless the interpretation is plainly errone-
ous or inconsistent with the directive.” Siskiyou Reg’l Educ.
Project v. U.S. Forest Serv., 565 F.3d 545, 554-55 & n.9 (9th
Cir. 2009) (citing Auer v. Robbins, 519 U.S. 452, 461 (1997)).
But the Service’s interpretation of the elk-cover requirement
in the Plan is plainly erroneous. The Plan requires that the
two-thirds cover be maintained “over time.” It further pro-
vides that “[s]ubsequent timber sale activity will be allowed
after regeneration provides hiding cover.” The Service’s inter-
pretation would allow iterative Service actions to whittle elk
cover down to nearly nothing so long as each individual
action removed only 33% of then-existing cover.

[23] Alternatively, the Service argues that even if the Proj-
ect violates the Gallatin Plan’s elk-cover requirement, the
error is harmless given the large elk populations in the Project
area. But “[i]t is well-settled that the Forest Service’s failure
to comply with the provisions of a Forest Plan is a violation
of NFMA.” Native Ecosystems Council, 418 F.3d at 961. “If
the Forest Service thinks any provision . . . of the Plan is no
longer relevant, the agency should propose amendments to the
. . . Plan altering its standards, in a process complying with
NEPA and NFMA.” Id. Although current elk populations may
meet or exceed Montana objectives, those objectives cannot
replace federal management objectives. The Service’s own
research scientists have written, in guidelines for elk manage-
ment, “Reducing habitat effectiveness should never be consid-
ered as a means of controlling elk populations. A population
over target is not a Forest Service habitat problem.”

[24] We therefore conclude that the Service has violated
the Gallatin Plan, and NFMA, by not ensuring that the Project
 HAPNER v. TIDWELL 14189
complies with the current Gallatin Plan elk-cover require-
ment. We remand to the Service to remedy this error.

Conclusion

We affirm the district court’s grant of summary judgment
to the Service in almost all respects. We remand, however,
because the Project fails to comply with the Gallatin Plan’s
elk-cover requirements.

AFFIRMED in part, REVERSED in part, and
REMANDED. Each side to bear its own costs.