Kurt Meister v. U.S. Department of Agriculture

RECOMMENDED FOR FULL-TEXT PUBLICATION
Pursuant to Sixth Circuit Rule 206
File Name: 10a0318p.06

UNITED STATES COURT OF APPEALS
FOR THE SIXTH CIRCUIT
_________________

X
-
KURT JAY MEISTER,
-
Plaintiff-Appellant,
-
-
No. 09-1712
v.
,
>
-
-
U.S. DEPARTMENT OF AGRICULTURE and

Defendants-Appellees. -
UNITED STATES FOREST SERVICE,
-
N
Appeal from the United States District Court
for the Eastern District of Michigan at Detroit.
No. 07-13008—Gerald E. Rosen, Chief District Judge.
Argued: March 8, 2010
Decided and Filed: September 29, 2010
Before: MERRITT, COOK, and KETHLEDGE, Circuit Judges.

_________________

COUNSEL
ARGUED: Kurt J. Meister, Novi, Michigan, for Appellant. Brian C. Toth, UNITED
STATES DEPARTMENT OF JUSTICE, Washington, D.C., for Appellee. ON BRIEF:
Kurt J. Meister, Novi, Michigan, for Appellant. Brian C. Toth, UNITED STATES
DEPARTMENT OF JUSTICE, Washington, D.C., for Appellee.
_________________

OPINION
_________________

KETHLEDGE, Circuit Judge. An agency is not entitled to deference simply
because it is an agency. It is true that agencies are more specialized than courts are. But
for courts to defer to them, agencies must do more than announce the fact of their
comparative advantage; they must actually use it. And that means, among many other

1
No. 09-1712 Meister v. U.S. Dep’t of Agric., et al. Page 2

things, that the agency must apply—rather than disregard—the relevant statutory and
regulatory criteria.

Kurt Meister, a Michigan attorney appearing pro se, argues that the United States
Forest Service disregarded the relevant criteria here. Specifically, he claims that the
Service failed to comply with several of its own regulations and one federal statute in
developing its 2006 management plan for the Huron-Manistee National Forests in
Northern Michigan. For the most part, we agree with him; and to that extent we reverse
the district court’s entry of judgment in the Service’s favor and remand the case so that
the Service may comply with those requirements forthwith.

I.

A.

This case concerns the Service’s management of recreational activities in the
Huron-Manistee National Forests (the “Forests” or “Forest”). The Forests occupy about
970,000 acres on each side of the northern one-third of Michigan’s Lower Peninsula.
In the east, the Huron National Forest ranges between 12 and 30 miles long from north
to south, and stretches 60 miles wide from west to east, reaching the shores of Lake
Huron. In the west, the Manistee National Forest is about 75 miles long and 40 miles
wide, reaching Lake Michigan near Manistee. The Forests serve a variety of interests,
including recreation, timber harvesting, and wildlife habitat. They lie within a two-hour
drive of 7.4 million residents of Michigan, and draw visitors from beyond the State.
Although technically separate, the Forests have been managed as a single unit since
1945.

1.

Every national forest is subject to a “land and resource management plan[.]”
16 U.S.C. § 1604(a). The National Forest Management Act requires each forest’s plan
to be revised every fifteen years, id. § 1604(f)(5), but in practice the interval often
stretches to twenty. Developing a plan is a formidable process: “The Service must
develop its management plans in conjunction with coordinated planning by a specially-
No. 09-1712 Meister v. U.S. Dep’t of Agric., et al. Page 3

designated interdisciplinary team, extensive public participation and comment, and
related efforts of other federal agencies, state and local governments, and Indian tribes.”
Sierra Club v. Marita, 46 F.3d 606, 609 (7th Cir. 1995) (citing 36 C.F.R. §§ 219.4-
219.7). The Service’s own regulations prescribe in great detail the procedures the
Service must follow in developing a forest plan. Those regulations make clear, as the
Service itself states in this appeal, that the Service must balance competing uses of the
Forests. See 36 C.F.R. § 219.4(a), (b); 16 U.S.C. § 1604(e)(1). (We cite the 2000
version of the regulations throughout, which is what the Service used in developing the
Plan.) Under the National Environmental Policy Act, the Service must also discuss a
range of alternative plans and assess the environmental impact of the alternative that it
proposes to adopt. 42 U.S.C. § 4332(2)(C)(i), (iii).

Once the development process is complete, a proposed final plan and final
environmental impact statement “are sent to the Regional Forester, who directs one of
four national forest regions, for review.” Marita, 46 F.3d at 609. If the Regional
Forester approves the plan and the impact statement, he issues both documents along
with a Record of Decision explaining his reasoning. The approved plan and impact
statement may be appealed to the Chief of the Forest Service, whose ruling becomes a
final administrative decision if the Secretary of Agriculture chooses not to review it.
36 C.F.R. §§ 219.10(d), 211.18.

2.

The Service issued a management plan for the Forests in 1986. In 2003, the
Service published a notice of intent to revise the plan. The Service thereafter held public
meetings and solicited public comments as to how to revise the plan. In 2005, the
Service published a draft environmental impact statement that described three alternative
plans and designated a preferred one. After receiving additional public comments for
three months, the Service prepared a final plan (the “Plan”) and a final environmental
impact statement for the Forests. In a Record of Decision dated March 2006, the
responsible Regional Forester approved both documents. See 71 Fed. Reg. 30399.
No. 09-1712 Meister v. U.S. Dep’t of Agric., et al. Page 4

3.

Meister commented on the Plan throughout its development. Those comments
reveal fluency with the language of the relevant statutes and regulations; and they
explained in considerable detail why Meister thought the Service was not meeting its
obligations under the law. His principal comment was that, in developing the Plan, the
Service had disregarded certain processes prescribed in its own regulations, so as to
favor gun hunters and snowmobile users over other persons—for example, hikers and
birdwatchers—who use the Forests for quiet, solitary activities. He also commented
that the Service should close more areas of the Forests to motorized activity than the
Service seemed likely to close in the Plan. It appears that the Service disagreed with all
of Meister’s comments.

Meister thereafter appealed the Plan administratively. The Service Chief issued
a decision that upheld the Plan in all but one respect: the Service was prohibited from
expanding snowmobiling to all open, unplowed roads in the Forests. The Secretary of
Agriculture declined to review the Reviewing Officer’s decision.

Meister then filed suit in the district court, challenging the Plan under the
Administrative Procedures Act. See 5 U.S.C. § 706. He and the Service each filed
motions for summary judgment. The district court granted the Service’s motion, holding
in general terms that the Service had complied with the applicable regulations. The
district court denied Meister’s motion.

This appeal followed.

II.

A.

The Service does not argue before us that Meister lacks standing to assert his
claims. But standing and ripeness have proved uncertain in cases like this one. See,
e.g., Ohio Forestry Ass’n, Inc. v. Sierra Club, 523 U.S. 726 (1998). So we pause to
address those issues here.
No. 09-1712 Meister v. U.S. Dep’t of Agric., et al. Page 5

Standing has three elements. “First, the plaintiff must have suffered an ‘injury
in fact’—an invasion of a legally protected interest which is (a) concrete and
particularized, and (b) actual or imminent, not conjectural or hypothetical.” Lujan v.
Defenders of Wildlife, 504 U.S. 555, 560 (1992) (internal citations and quotation marks
omitted). Second, the injury must be “fairly traceable to the challenged action of the
defendant.” Id. (internal alterations omitted). Third, it must be likely that the injury will
be “redressed by a favorable decision.” Id. at 561.

The Supreme Court has “held that environmental plaintiffs adequately allege
injury in fact when they aver that they use the affected area and are persons for whom
the aesthetic and recreational values of the area will be lessened by the challenged
activity.” Friends of the Earth, Inc. v. Laidlaw Envtl. Servs. (TOC), Inc., 528 U.S. 167,
183 (2000) (internal quotation marks omitted). Here, Meister alleges that he uses the
Forests and that the Service’s failures to follow certain procedures in developing the
Plan have resulted in more widespread gun hunting and snowmobile usage than there
would be had the Service complied with the procedures. Consequently, Meister
contends, the Service’s failures have “substantially diminishe[d] Plaintiff’s walking,
hiking, mountain biking, kayaking, cross country skiing, and snowshoeing experiences”
in the Forests. Per Friends of the Earth, that contention states an injury in fact.

There remains the question of ripeness. The doctrine seeks “avoidance of
premature adjudication”; and, in cases involving challenges to agency action, ripeness
depends on whether the “administrative decision has been formalized and its effects felt
in a concrete way by the challenging parties. Ohio Forestry, 523 U.S. at 732-33
(internal quotation marks omitted).

In Ohio Forestry, the Sierra Club sought to challenge provisions of a forest-
management plan that designated certain areas as open to logging. The Supreme Court
did not question that the aesthetic harm alleged by the Club would amount to an injury
in fact. But the Court observed that the Plan itself “does not give anyone a legal right
to cut trees, nor does it abolish anyone’s legal authority to object to trees being cut.” Id.
at 733. To the contrary, additional agency action—namely, issuance of a site-specific
No. 09-1712 Meister v. U.S. Dep’t of Agric., et al. Page 6

permit—was required before anyone could engage in the logging that the Sierra Club
said would harm its interests. Thus, the Court held, the plan had not yet “inflict[ed]
significant practical harm upon the interests that the Sierra Club advance[d.]” Id. Hence
the case was not ripe.

Meister’s case is different. Unlike logging, the activities about which Meister
complains—gun hunting and snowmobile use—do not require further action by the
Service before they can occur. To the contrary, they have in fact occurred ever since the
Plan’s issuance, with the resultant harms that Meister now alleges. Thus, the Plan itself
has harmed him in concrete ways. His claims are ripe.

B.

We turn next to our standard of review. As an initial matter, we do not defer to
the district court’s decision, but instead review the administrative decision as if we were
the first reviewing court. Schuck v. Frank, 27 F.3d 194, 197 (6th Cir. 1994). Our review
of the agency’s decision is governed by the Administrative Procedure Act. The APA’s
relevant section provides in relevant part:

The reviewing court shall—

(2) hold unlawful and set aside agency action, findings, and
conclusions found to be—
(A) arbitrary, capricious, an abuse of discretion, or
otherwise not in accordance with law; [or] . . .
(D) without observance of procedure required by law[.]
5 U.S.C. § 706.

Meister’s claims arise in part under § 706(2)(D), since he contends that the
Service failed to follow certain procedures in implementing the Plan. Under that
subsection we review agency action de novo. Coalition for Gov’t Procurement v. Fed.
Prison Indus., Inc., 365 F.3d 435, 457 (6th Cir. 2004). But even in cases arising under
§ 706(2)(D), our review as a practical matter is often more deferential than that. The
reason is that the question whether a certain procedure is required in a particular
No. 09-1712 Meister v. U.S. Dep’t of Agric., et al. Page 7

circumstance, or whether a certain methodology satisfies the procedure, is often left to
the agency’s discretion. See, e.g., Save Our Cumberland Mountains v. Kempthorne, 453
F.3d 334, 339 (6th Cir. 2006) (stating that we review an agency’s “decision that an
environmental impact statement need not be prepared, under the deferential ‘arbitrary
and capricious’ standard”); Hughes River Watershed Conservancy v. Johnson, 165 F.3d
283, 289 (4th Cir. 1999) (“[a]gencies are entitled to select their own methodology as
long as that methodology is reasonable”). So even in cases arising under § 706(2)(D),
the arbitrary-and-capricious standard frequently governs.

Meister’s claims also arise to some extent under § 706(2)(A), since he contends
that certain aspects of the Plan were arbitrary. As a general matter, agency action is
arbitrary or capricious if

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

Motor Vehicle Mfrs. Ass’n of U.S., Inc. v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29,
43 (1983).

Applying these standards, we turn to the merits of Meister’s claims.

III.

A.

“It is an elemental principle of administrative law that agencies are bound to
follow their own regulations.” Wilson v. Comm’r of Soc. Sec., 378 F.3d 541, 545 (6th
Cir. 2004). Meister claims that the Forest Service violated three of its own procedural
regulations in developing the Plan.
No. 09-1712 Meister v. U.S. Dep’t of Agric., et al. Page 8

1.

Meister first claims that the Forest Service failed to comply with § 219.21(a)(2)
of its regulations when it developed the Plan. That subsection provides:

(a) Forest planning shall identify—
...
(2) The recreational preferences of user groups and the settings
needed to provide quality recreation opportunities[.]

The parties disagree about what this provision means. The Forest Service reads
“recreational preferences” to mean “activities,” so that this regulation requires
“essentially, a demand/supply analysis.” Gov’t Br. at 15. Under this interpretation, the
Service must estimate the demand for various activities in the Forest—e.g., camping,
cross-country skiing, and snowmobiling—and then assess the quantity of Forest lands
“needed” to meet the demand for each activity.

Meister disagrees, arguing that a different provision, § 219.21(b), requires a
demand-supply analysis. (It does, but of a type not relevant here—namely, for
“developed recreational facilities[.]”) Under Meister’s interpretation, the identification
required by subsection (a)(2) would lay the foundation for a demand-supply analysis,
rather than constitute the analysis itself. Specifically, he reads “recreational preferences
of user groups” to mean not the users’ preferred activities themselves, but rather what
those users seek to obtain from those activities; as, for example, “cross-country skiers
desire a quiet recreation experience.” Meister Reply at 7-8. So in Meister’s view, the
Service must determine the kinds of physical settings in which various users would
prefer to engage in their activities, rather than merely the settings in which those
activities can occur. Thus, for example, the Service should determine that cross-country
skiers, for a variety of reasons, do not want their trails located in close proximity to
snowmobile trails. Meister’s ultimate concern seems to be that the Service’s
interpretation allows it to consider subjectively unsatisfactory settings, rather than the
users’ preferred ones, in determining whether an adequate supply of Forest lands exists
for a particular activity.
No. 09-1712 Meister v. U.S. Dep’t of Agric., et al. Page 9

We defer to an agency’s plausible interpretation of its own regulation. Auer v.
Robbins, 519 U.S. 452, 461 (1997). “Recreational preferences,” as used in
§ 219.21(a)(2), can just as easily be read to mean the activities that users prefer, as it can
the physical settings in which they prefer to do them. And though not as easily, “settings
needed” can be read to mean the quantity of lands necessary to support an activity. So
the Service’s interpretation of this subsection lies within its discretion; and thus the
subsection does require essentially a demand-supply analysis. But the question remains:
Supply of what? The regulation provides a straightforward answer: The supply of lands
“needed to provide quality recreation opportunities[.]” 36 C.F.R. § 219.21(a)(2)
(emphasis added). It is not enough, therefore, for the Service merely to identify the
supply of lands on which an activity can occur. It must instead identify the supply of
lands on which participants in that activity are afforded a “quality recreation
opportunit[y.]” So Meister has a point after all.

With this interpretation in hand, we consider the substance of Meister’s claim.
He has two complaints. First, he says the Service failed to identify “the settings needed
to provide quality recreation opportunities for hikers, backpackers, and cross-country
skiers[.]” Meister Br. at 18. The Service responds that a methodology it calls the
“Recreation Opportunity Spectrum” (“ROS”) accomplishes that identification for those
activities and many more. The ROS classifies forest lands into seven types, depending
on their degree of development. The classifications are: Primitive, Semiprimitive
Nonmotorized, Semiprimitive Motorized, Roaded Natural, Rural, and Urban. ROS
Users Guide at 2. The ROS contains a detailed description of each classification.

“The Service is entitled to use its own methodology, unless it is irrational.”
Marita, 46 F.3d at 621; see also California v. Watt, 712 F.2d 584, 597 (D.C. Cir. 1983).
The ROS easily passes that test: It is not an irrational, but a thoughtful methodology for
matching settings and activities, among other planning purposes. Suffice it to say that
we agree with the Service on this point.

But Meister’s other complaint is more problematic. He contends that, as a part
of its demand-supply analysis, the Service significantly overestimated snowmobile use
No. 09-1712 Meister v. U.S. Dep’t of Agric., et al. Page 10

and underestimated cross-country ski activity in the Forests. The issue begins with the
fact that the Service’s methodology for measuring the intensity of different activities in
the Forests—the “National Visitor Use Monitoring” (“NVUM”) survey—indicated that
no visitors to the Forests reported snowmobiling or cross-country skiing as the primary
purpose of their visit in the year 2000. Recreational Supply and Demand Analysis (Oct.
2004) (“Demand Analysis”) at 11. “[B]ased upon discussions with Dr. Daniel Stynes,
Professor, Michigan State University[,]” however, and the “professional judgment of the
Huron-Manistee’s staff and management,” the Service adjusted its estimate of annual
snowmobile visits from zero to 120,000 for the year 2000 and to 138,000 for the year
2010. Final Environmental Impact Statement (March 2006) (“Final Statement”) at III-
337, 287; Demand Analysis at 21. But the Service made no parallel adjustment for
cross-country visitors, notwithstanding its recognition that demand “for this activity is
projected to increase approximately 50 percent over the next 50 years[,]” a rate “faster
than the rate of population growth.” Demand Analysis at 30, 23. The estimates of cross-
country visitors, for the years 2000 to 2050, were left at zero.

Meister says this treatment of the numbers amounts to “everything imaginable
short of making up the data about snowmobile use.” Meister Br. at 17 n.4. The Service
responds that it “documented its methodology” in the Demand Analysis and that
“[c]ommunications with the Michigan State researcher and internal discussions about
his opinions were also well-documented in the administrative record.” Gov’t Br. at 22.

The whole episode debases the coinage of agency deference. We begin with the
discussions with Dr. Stynes. So far as the record is concerned, the discussions are in fact
nothing more than an single email exchange comprising several pages. The exchange
is informal—Dr. Stynes himself says at the outset that “I’m not ready to present formal
estimates for a peer review”—and the exchange actually focuses almost entirely on Dr.
Stynes’ belief that the total number of annual visits (as opposed to just snowmobile
visits) to the Forests should be adjusted from 1.3 million (the number suggested by the
NVUM) to 3 million. A Forest Service Social Scientist, Dr. Donald English, expressed
serious concerns about that proposed adjustment, stating that “I reiterate that I think it
No. 09-1712 Meister v. U.S. Dep’t of Agric., et al. Page 11

would require a fair bit of work to justify using some other visitation estimate other than
the NVUM estimate.” So far as we can tell, no further work was done—another Service
official said simply to choose either number, “[j]ust make sure it’s something
defensible”—and the Service chose the estimate of 3 million. But the validity of that
number is not before us today.

What is before us is the snowmobile number, which as noted above was adjusted
from zero to 120,000 for the year 2000. According to the Service’s Demand Analysis,
“Stynes estimated that snowmobile visits on the National Forest were approximately
120,000” for that year. Demand Analysis at 11. But that characterization is unfair to Dr.
Stynes: He made no such estimate. What he does say—in a passage comprising three
sentences—is that the NVUM survey was surely wrong in estimating zero snowmobile
visits to the Forest, that a earlier State study estimated about 1.2 million snowmobile
user-days in the entire northern lower peninsula of Michigan in 1996, and that if “10%
of the activity is on the Forests, we get 120,000 [National Forests] snowmobile visits.”
There is nothing more. Those three sentences, so far as we can tell from the record, are
the entire basis of the Service’s adjustment of the snowmobile estimate from zero to
120,000.

So we see two serious problems with the Service’s estimates of snowmobile and
cross-country visitors. The first is that there is scarcely any basis for the snowmobile
estimate. Behind all of the Service’s invocations of communications, discussions, and
professional judgment, there lie only those three sentences. Those sentences themselves
contain not an estimate, but a hypothetical. The record otherwise contains not a hint of
professional judgment on the Service’s part as to why the 120,000 number, as opposed
to some other number, is reasonable. We mean no criticism of Dr. Stynes, since in this
respect his email was put to a use that he appears not to have contemplated. But we do
say that the nation’s forests deserve better than this as the basis for significant decisions
regarding their management.

The second problem is the disparate treatment of the cross-country estimate.
Most, if not all, of the Service’s rationales for boosting the snowmobile numbers
No. 09-1712 Meister v. U.S. Dep’t of Agric., et al. Page 12

generally (as opposed to pegging them at 120,000 specifically)—namely, “abnormally
low snowfall” during the survey year, high gas prices, and that participants in the activity
typically enter the Forests via trails rather than at developed sites where the survey forms
were located—apply fully to cross-country visitors. Meister quite reasonably asks why
snowmobile visits received an upward adjustment, but cross-country visits did not. The
Service responds that there was “[i]nsufficient data” from which to make an estimate of
cross-country visitors. Given the data found sufficient for the snowmobile estimate, that
response is hardly persuasive. And given that the Demand Analysis contains reliable
(and likely peer-reviewed) data showing the participation rate in cross-country skiing to
be about half that for snowmobiling in the parts of the nation near the Forests, see
Demand Analysis at 23, it would seem that once the Service had a snowmobile estimate
in hand, a cross-country estimate would soon follow. But there is none: So far as the
tables in the Demand Analysis are concerned, the Forests have seen their last cross-
country skier.

Deference must be earned. Kadia v. Gonzales, 501 F.3d 817, 821 (7th Cir.
2007). It was not earned in the two respects that we discuss here. The Service’s
estimates of snowmobile and cross-country visitors to the Forests are entirely arbitrary.
And—under the Service’s own interpretation of § 219.21(a)(2)—those estimates are
integral to the demand-supply analysis required under that subsection. To that extent,
the Service’s issuance of the Plan was arbitrary and without observance of procedure
required by law. 5 U.S.C. § 706(2)(A), (D).

2.

Meister also claims that, in developing the Plan, the Service did not coordinate
with Michigan’s recreational planning as required by § 219.21(e). That subsection
provides that the Service’s planning efforts

shall be coordinated to the extent feasible with present and proposed
recreation activities of local and State land use or outdoor recreation
plans, particularly the State Comprehensive Outdoor Recreation Plan,
and recreation opportunities already present and available on other public
No. 09-1712 Meister v. U.S. Dep’t of Agric., et al. Page 13

and private lands, with the aim of reducing duplication in meeting
recreation demands.

36 C.F.R. § 219.21(e). In plainer English, the subsection requires the Service, “to the
extent feasible, to coordinate with the State in eliminating duplicate recreation
opportunities on state and national forests.” Gov’t Br. at 16. Plainer still: The acreage
already dedicated to a particular activity on state land should be considered in deciding
how much federal land to allocate to the activity.

Meister claims that the Service failed to coordinate in this manner with respect
to gun hunting and snowmobiling. The first step in this coordination, it would seem,
would be simply to determine how much land is already available for these activities on
state lands. The Service has not done that. In its Answer to Meister’s Complaint, the
Forest Service admitted—long after the Plan had been approved—that it “lack[s]
sufficient knowledge or information to respond to the allegations concerning firearm
hunting on state forest lands managed by the State of Michigan[.]” The Service made
essentially the same admission with respect to snowmobiling on state lands. It is hard
to see how the Service could even attempt to reduce duplication of hunting and
snowmobiling opportunities on state lands without that basic information.

Meister also seeks to show actual duplication as a means of proving that the
Service made no attempt to reduce it. Specifically, he notes—and the Service does not
dispute—that “[h]unting is permitted on approximately 3.8 million acres of state forest
land and over 340,000 acres of state game and wildlife areas in the State of Michigan[,]”
and that “there are more than 6,500 miles of snowmobile trails in the State[.]” Meister
Br. at 14, 15. He further notes that hunting is permitted virtually everywhere in the
Forests, save perhaps the parking lots—meaning that the Forests’ duplication of state-
land hunting opportunities, to the extent there is any within the meaning of the rule,
literally could not be greater. And he asks specifically why it was not feasible to reduce
duplication by prohibiting gun hunting (as opposed to bow hunting, which is largely
silent, and to which Meister seems not to object) and snowmobiling in at least the
Primitive and Semiprimitive Nonmotorized areas of the Forest. Those areas, as defined
No. 09-1712 Meister v. U.S. Dep’t of Agric., et al. Page 14

by the ROS, are supposed to bring an “extremely high probability” and a “high
probability,” respectively, “of experiencing isolation from the sights and sounds of
humans.” ROS Users Guide, Table 1 (“Experience Characterization”). Firearm and
snowmobile use, as the Service itself recognizes, involve high-intensity noises created
by humans. Primitive and Semiprimitive areas together comprise less than 66,000 acres
of the 970,000 acres occupied by the Forests, or just 6.75% of the Forests’ acreage. See
Final Statement at III-275. Meister therefore contends that the Service’s failure to
restrict these activities in even these areas—particularly in light of the circa 4.1 million
acres open to hunting on state land, and the 6,500 miles of snowmobile trails there—is
strong evidence that the Service did not comply with the coordination procedures
mandated by § 219.21(e) so far as gun hunting and snowmobiling are concerned.

The Service’s response is insubstantial. First, it says that it uses the State
Comprehensive Outdoor Recreation Plan as a tool to monitor recreation demand. But
that only means the Service has done half of what its regulation requires.
Section 219.21(e), as shown above, requires consideration of the extent to which the
supply of state lands already meets that demand. And the Service admits it lacks
knowledge on that point. Second, the Service again tells us that it consulted with Dr.
Stynes, and that its “communications” with him and “internal discussions” regarding his
opinions are “well-documented.” Gov’t Br. at 25. We are referred to the same email
exchange. The exchange has even less to say about duplication (that is, nothing) than
it does about snowmobile visits. Finally, the Service cites a letter from the Michigan
Department of Natural Resources, in which the State says that the Plan’s increases in
Semiprimitive Nonmotorized and Semiprimitive Motorized areas “work[] to meet the
demand” for those areas and “may supplement existing inter-agency recreational efforts
that address recreation demand in a cooperative manner (i.e. ORV trails, pathways,
campground placement).” Final Statement at J-157-58. The State’s comments speak
well of the Service’s decision to increase Semiprimitive acreage in the Forests. But they
simply have nothing to do with whether hunting and snowmobiling opportunities on
Forest lands duplicate those on state lands.
No. 09-1712 Meister v. U.S. Dep’t of Agric., et al. Page 15

Thus, with respect to hunting and snowmobiling, there is no evidence that the
Service complied with § 219.21(e), and ample evidence that it did not. In that respect
as well, therefore, the Plan’s approval was “without observance of procedure required
by law.” 5 U.S.C. § 706(2)(D).

3.

Meister’s remaining claim under the Management Act is that the Service failed
to comply with 36 C.F.R. § 219.21(g). That regulation provides:

Off-road vehicle use shall be planned and implemented to protect land
and other resources, promote public safety, and minimize conflicts with
other uses of the National Forest System lands. Forest planning shall
evaluate the potential effects of vehicle use off roads and, on the basis of
the requirements of 36 CFR part 295 of this chapter, classify areas and
trails of National Forest System lands as to whether or not off-road
vehicle use may be permitted.

This regulation plainly contemplates that off-road vehicle use may conflict with
other interests and uses of the Forests. In Meister’s view, the regulation means that, “if
there is a conflict between off-road vehicles and other uses, off-road vehicles lose.”
Meister Br. at 19. The Service disagrees. We agree with the Service: The regulation
requires it only to “minimize conflicts” between off-road vehicles and other uses, not to
eliminate them.

But the mandate to minimize remains. Meister contends it was not met. He is
concerned about snowmobiles in particular; and the basis of his claim is that—as the
Service itself states— snowmobile activities “have continued to occur in or near several
existing semiprimitive nonmotorized management areas.” Final Statement at III-313.
Those areas, as noted above, are supposed to be characterized by “a large probability of
isolation from the sights and sounds of others.” Id. at III-312. Therein lies the conflict;
and Meister suggests that its elimination was feasible enough to fall within the mandate
to minimize.

The Service offers several responses. First, it says that “the forest plan is not an
appropriate vehicle for making individualized trail closure decisions.” Gov’t Br. at 26.
No. 09-1712 Meister v. U.S. Dep’t of Agric., et al. Page 16

That is true so far as it goes; once the Service decides that an area should generally be
open to off-road vehicle usage, the decision whether to close trails within it can properly
be made on an individualized basis at the field level. But § 219.21(g) by its plain terms
imposes the bulk of its obligations at the planning stage; and one such obligation, as we
read the regulation, is to determine whether certain classes of areas and trails ought to
be altogether off-limits to off-road vehicle use. Meister says that one such class are trails
that the Service itself admits are “in or near” Semiprimitive Nonmotorized areas. That
claim is properly presented at the Plan level.

The Service’s other responses go to the merits. One is that “[s]ome of the
nonconformities result because the snowmobile trail coincides with a county road that
the Service does not have jurisdiction to close.” Gov’t Br. at 27. That is a good answer
as to those trails. Less good is that other trails (which we call “pre-designation trails”)
“existed prior to their designation or proposal for designation.” Final Statement at III-
313. The argument seems to be that these trails were used by off-road vehicles before
being designated for that use, and thus should be grandfathered in. But elsewhere in the
Plan documents, the Service repeatedly emphasizes the importance of limiting off-road
vehicle usage to “designated routes only.” Record of Decision at 10. So off-road
vehicle use on undesignated trails is a bad thing; and thus it hardly seems rational to say
that a trail should remain open to off-road vehicles simply because the trail was used by
them before its designation. And it is likewise incoherent to say—as a basis for keeping
other trails (“club trails”) open to off-road vehicles—that “[s]nowmobile clubs use
motorized equipment to groom snowmobile trails on roads within several of the current
semiprimitive nonmotorized management areas.” Final Statement at III-313. The mere
fact of a nonconformity is no reason to allow it to continue.

There conceivably might be reasons for keeping pre-designation and club trails
open to off-road vehicle usage—though presumably the reasons should be good ones,
given the Service’s own recognition that these uses are “nonconformities.” If the
Service were to articulate good reasons, we would defer to them. As the record now
stands with respect to these trails, however, the Service has not “articulate[d] a
No. 09-1712 Meister v. U.S. Dep’t of Agric., et al. Page 17

satisfactory explanation for its action including a rational connection between the facts
found and the choice made.” Motor Vehicle Mfrs. Ass’n of U.S., 463 U.S. at 43 (internal
quotation marks omitted).

Thus, as matters now stand, the Service has not complied with § 219.21(g)’s
mandate to minimize conflicts between off-road vehicle use and other uses and interests
of the Forests. To that extent, the Plan’s approval was arbitrary and without observance
of procedure required by law. 5 U.S.C. § 706(2)(A), (D).

B.

Meister’s remaining two claims arise under the National Environmental Policy
Act. The Act “serves procedural rather than substantive goals.” Save Our Cumberland
Mountains, 453 F.3d at 338. The Act has a specialized standard of review for
arbitrariness: “In deciding whether the agency acted arbitrarily, we will not substitute
our own judgment” for that of the agency, “but we will insist that the agency has, in fact,
adequately studied the issue and taken a hard look at the environmental consequences
of its decision.” Id. at 339 (internal quotation marks omitted).

The Act requires, among other things, that an agency “study, develop, and
describe appropriate alternatives to recommended courses of action in any proposal
which involves unresolved conflicts concerning alternative uses of available resources.”
42 U.S.C. § 4332(2)(E). Meister argues that there were unresolved conflicts between
different recreational uses of the Forests here, and that the Service should have, but did
not, consider his proposed alternatives to resolve them.

1.

Meister’s first claim is based upon a conflict between the high-intensity noise
generated by gun hunting and snowmobile use, on the one hand, and the quiet
recreational activities that Primitive and Semiprimitive Nonmotorized areas are
supposed to support, on the other. He argues that the Service should have, but did not,
consider the alternative of closing these areas to gun hunting and snowmobiling as a
No. 09-1712 Meister v. U.S. Dep’t of Agric., et al. Page 18

means of resolving this conflict. Meister proposed this alternative during the
administrative process, but the Service rejected it out of hand.

“As a general matter, the range of alternatives that must be discussed under the
National Environmental Policy Act is a matter within an agency’s discretion.” Save Our
Cumberland Mountains, 453 F.3d at 342 (internal quotation marks omitted). Thus, an
“agency may apply a ‘rule of reason’ in this area and discuss only ‘reasonable’
alternatives to the proposed action.” Id. at 346. The question, then, is whether the
Service acted within its discretion in rejecting Meister’s alternative without any
discussion.

That rejection was based on a series of factual and legal errors. The principal
factual error is that, during both the administrative and now the legal process, the Service
has treated Meister’s proposed restriction on hunting as something it obviously is not:
Namely, a proposal that the Service impose “[a] complete ban on hunting in the
Forests[.]” Gov’t Br. at 31; see also May 27, 2007 letter from Forest Supervisor Leanne
Marten to Kurt Meister (“it is my determination that a general closure” of the Forests
to hunting “is not justifiable”); Final Statement at J-93 (same).

All but one paragraph of the government’s briefing of this claim is dedicated to
ruling out such a complete ban. (The Service did accurately characterize Meister’s
proposal in its decision rejecting this part of his administrative appeal, but in doing so
merely adopted Supervisor Marten’s complete-ban rationale.) But a complete ban is not
what Meister proposes. He proposes that gun hunting and snowmobiling be prohibited
only in Primitive and Semiprimitive Nonmotorized areas (the “challenged areas”), which
as noted above comprise only 6.75% of the Forests. See, e.g., Meister Br. at 21 (arguing
that the Service “impermissibly failed to consider alternatives to a general closure such
as only closing Wilderness and Semi-Primitive Non-Motorized areas to hunting and
snowmobiling”); Feb. 28, 2003 letter from Kurt Meister to James Schuler (stating that
gun hunting and snowmobiling “should be banned in all Primitive and Semi-Primitive
Non-Motorized areas”). And the Service has compounded (no pun intended) its error
by overlooking that Meister proposes to restrict only gun hunting, not bow hunting.
No. 09-1712 Meister v. U.S. Dep’t of Agric., et al. Page 19

These aspects of Meister’s proposal are hardly abstruse; and they narrow the
proposal’s scope both qualitatively (so that it concerns only gun hunting rather than
bow) and quantitatively (so that it concerns only 6.75% of the Forests’ acreage rather
than 100%). An agency cannot reject a proposed alternative by treating it as something
it is not. To exercise discretion with respect to a proposed alternative, the agency must
treat the alternative on its own terms. The Service has not done that here.

In its brief to this court, the Service does briefly discuss the more narrow closure
(although it still overlooks the gun versus bow distinction). In doing so the Service
makes different contentions with respect to snowmobiles and gun hunting. As to the
former, the Service essentially contends that Congress did not require it to close the
challenged areas to snowmobiling. But Congress did not prohibit such a closure either;
and the mere fact that Congress left the adoption of that alternative to the agency’s
judgment does not mean that the agency was free not to consider the alternative at all.
The National Environmental Policy Act and its regulations say precisely the contrary.

As to gun hunting, the Service contends that Congressional policy favors leaving
the entire Forest open to hunting, so that even the narrow closure that Meister proposes
is beyond its authority. Specifically, the Service notes that the Multiple-Use Sustained-
Yield Act of 1960, 16 U.S.C. § 528, the Wilderness Act, 16 U.S.C. § 1133(d)(7), and the
Federal Land Policy and Management Act, 43 U.S.C. § 1732(b), each make clear that
they do not affect “the jurisdiction or responsibilities of the several States with respect
to wildlife and fish on the national forests.” 16 U.S.C. § 528. But there are clear
exceptions to that rule. The Federal Land Policy and Management Act expressly
provides that the Service “may designate areas of public land and of lands in the
National Forest System where, and establish periods when, no hunting or fishing will be
permitted for reasons of public safety, administration, or compliance with provisions of
applicable law.” 43 U.S.C. § 1732(b) (emphasis added). The Service’s own guidelines
echo the point:

Hunting, fishing, and trapping of fish and wildlife and associated
practices on National Forest System lands are subject to State fish and
wildlife laws and regulations, unless one or both of the following apply:
No. 09-1712 Meister v. U.S. Dep’t of Agric., et al. Page 20

1. State fish and wildlife laws and regulations conflict with
Federal laws; or
2. State laws and regulations would permit activities that conflict
with land and resource management responsibilities of the Forest
Service or that are inconsistent with direction in forest plans.

Forest Service Manual § 2643.1 (emphasis added).

Gun hunting is inconsistent with the “direction in forest plans” as set forth in the
ROS descriptions of the challenged areas, since those areas are supposed to present little
chance of encountering noise by humans. The Service cannot expect us to defer to its
ROS descriptions when they support its decision (which we have done above), but then
to disregard those same descriptions when they conflict with its decision. And the
descriptions conflict with the decision here. Meister has cited precisely that conflict
throughout the administrative and now legal process. The question whether the
Service’s own specific descriptions of the challenged areas, on the one hand, should
prevail over the State’s much more general rules permitting gun hunting, on the other,
is one that warrants a careful answer under the National Environmental Policy Act. See
Save Our Cumberland Mountains, 453 F.3d at 347.

Meister’s alternative likewise warrants consideration on grounds of compliance
with applicable law. 43 U.S.C. § 1732(b). The Service, to its credit, forthrightly
recognizes in its brief that it “has been charged by Congress with balancing a variety of
competing interests in its land use planning.” Gov’t Br. at 28 (citing 16 U.S.C.
§ 1604(e)(1) and 36 C.F.R. § 219.1(a), (b)). But striking a balance typically involves
some give on each side. We think that should be especially true for participants in
activities that do not conform to the area descriptions in the ROS. One might at least
expect them not to have the run of the areas. And so, in striking a balance between
competing uses of the Forests, one might expect the Service seriously to consider
whether, say, birdwatchers in fall should be able to enjoy their pastime, in 6.75% of the
Forests, without ducking for the occasional gunshot. Or whether, in some corner of the
Forest—especially an ostensibly “nonmotorized” one—a snowshoer should be able to
walk a trail without hearing the whine of snowmobile engines. The Service is charged
No. 09-1712 Meister v. U.S. Dep’t of Agric., et al. Page 21

with balancing competing uses of the Forests, rather than favoring one or two uses above
all others. And if that balance requires closure of certain areas to certain activities,
Congress has granted the Service that authority. There is no lawful policy that ties the
Service’s hands in this regard.

“[A]n alternative within the ambit of an existing standard . . . generally may not
be abandoned without any consideration whatsoever.” Save Our Cumberland
Mountains, 453 F.3d at 347 (internal quotation marks omitted). The Service was
incorrect to conclude that Meister’s proposed alternative fell outside the ambit of the
relevant standards here. It seems more likely to us that the Service’s decision not to
balance these competing uses, and to disregard its own ROS descriptions, is what fell
outside the relevant standards. Meister’s proposed alternative deserves serious
consideration under the National Environmental Policy Act.

2.

Meister’s final claim is that the Service should have considered designating
substantially more areas as Semiprimitive Nonmotorized than it did in the Plan. Unlike
Meister’s other proposed alternative, the Service did consider something akin to this one,
but concluded that the alternative was not viable. It had rational reasons for that
conclusion. See, e.g., Final Statement at III-313 (“[a]fter reviewing such things as
Recreation Opportunity Spectrum, road densities, old growth and the management and
ownership of adjoining private lands, no new areas were identified for consideration as
semiprimitive management areas”). We defer to those reasons, and reject this claim.

IV.

We summarize our holdings today. First, the Service’s estimates of snowmobile
and cross-country visitors to the Forests are arbitrary. Thus, the Service has not
complied with § 219.21(a)(2)’s requirement of a demand-supply analysis.

Second, the Service has not complied with the requirement that it coordinate its
recreational planning with that of the State of Michigan with the aim (to the extent
No. 09-1712 Meister v. U.S. Dep’t of Agric., et al. Page 22

feasible) of “reducing duplication in meeting recreation demands” with respect to gun
hunting and snowmobiling. 36 C.F.R. § 219.21(e)(2000).

Third, the Service’s reasons for keeping pre-designation and club trails open to
snowmobile use are arbitrary. Thus, the Service has not complied with § 219.21(g)’s
mandate to minimize conflicts between off-road vehicle use and other uses and interests
of the Forests.

Fourth, the Service violated the National Environmental Policy Act when it
failed to consider whether to close Primitive and Semiprimitive Nonmotorized areas to
gun hunting and snowmobile use, as Meister has proposed.

Each of these failures was material to the Plan’s development. To that extent,
the Plan’s approval was arbitrary or without observance of procedures required by law.

Given that holding, we have authority to “set aside” the Plan. See 5 U.S.C.
§ 706(2)(A), (D). We choose not to exercise that authority today, but instead grant the
Service a reasonable time to adopt a plan that complies with the law. Ninety days from
the date of our mandate seems to us ample time for that compliance. The district court
may extend that period upon some showing that the court finds compelling; but in any
event the Service shall comply forthwith.

The district court’s judgment is reversed with respect to the claims summarized
in Part IV of this opinion. Meister is entitled to judgment on those claims to the extent
described in that Part; and the claims are remanded to the district court for further
proceedings consistent with this opinion. The district court’s judgment is otherwise
affirmed.