Center for Native Ecosystems v. Cables

FILED
United States Court of Appeals
Tenth Circuit

December 17, 2007
PUBLISH Elisabeth A. Shumaker
Clerk of Court
UNITED STATES COURT OF APPEALS

TENTH CIRCUIT

CENTER FOR NATIVE
ECOSYSTEMS; BIODIVERSITY
CONSERVATION ALLIANCE; and
FOREST GUARDIANS,

Plaintiffs - Appellants,
v. No. 06-1130
RICK CABLES, in his official
capacity as Regional Forester, Region
2, United States Forest Service; and
UNITED STATES FOREST
SERVICE,

Defendants - Appellees,

and

POLE MOUNTAIN CATTLEMEN’S
ASSOCIATION, an Unincorporated
Association, and its members; BATH
SISTERS, LLC, a Wyoming Limited
Liability Company; MARK EISELE;
WARREN LIFESTOCK, LLC, a
Wyoming Limited Liability Company;
PETER HANSEN; BONHAM
RANCH, LLC, a Wyoming Limited
Liability Company; C. C. DAVIS &
CO., LLC, a Wyoming Limited
Liability Company; FERGUSON
RANCH, INC., a Wyoming
Corporation; GARDNER BROS.;
WILLADSEN BROS.; QUARTER
CIRCLE F QUARTER CIRCLE LONE
TREE RANCH, INC., a Wyoming
 Corporation; WYOMING STOCK
GROWERS ASSOCIATION, a
nonprofit Wyoming Corporation on
behalf of its members; LARAMIE
COUNTY FARMERS UNION, a
nonprofit Wyoming Corporation on
behalf of its members; WYOMING
FARM BUREAU FEDERATION, a
nonprofit Wyoming Corporation on
behalf of its members; WYOMING
ASSOCIATION OF
CONSERVATION DISTRICTS, a
nonprofit Wyoming Corporation on
behalf of its members,

Defendant-Intervenors -
Appellees.
_____________________________

STATE OF WYOMING;
PACIFIC LEGAL FOUNDATION;
NATIONAL ASSOCIATION OF
HOME BUILDERS; AMERICAN
FOREST & PAPER ASSOCIATION,

Amici Curiae.

APPEAL FROM THE UNITED STATES DISTRICT COURT
FOR THE DISTRICT OF COLORADO
(D.C. NO. 04-cv-2409-PSF-OES)

McCrystie Adams (Neil Levine with her on brief), of Earthjustice, Denver,
Colorado, for Plaintiffs - Appellants

Mark R. Haag, Environment & Natural Resources Division, Department of
Justice, Washington, D.C., (Sue Ellen Wooldridge, Assistant Attorney General,

-2-
and David Shilton, Department of Justice, Washington, D.C.; Diane M. Conolly,
Deputy Regional Attorney, and Kenneth P. Pitt, General Attorney, Office of the
General Counsel, Department of Agriculture, Golden, Colorado, with him on the
brief), for Defendants - Appellees.

Karen Budd-Falen and Marc R. Stimpert, Budd-Falen Law Offices, LLC,
Cheyenne, Wyoming, for Defendant-Intervenor - Appellees Wyoming Association
of Conservation Districts.

Daniel B. Frank, Frank Law Office, P.C., Cheyenne, Wyoming, for Defendant-
Intervenor - Appellees Pole Mountain Cattlemen’s Association, Wyoming Stock
Growers Association, Wyoming Farm Bureau Federation, and Laramie County
Farmers Union.

Thomas R. Lundquist, J. Michael Klise, Crowell & Moring, LLP, Washington,
D.C., and Duane J. Desiderio, Staff Vice President, Legal Affairs, National
Association of Home Builders, Washington, D.C., filed an amicus curiae brief for
the National Association of Home Builders and American Forest & Paper
Association.

Patrick J. Crank, Attorney General, and Thomas W. Rumpke, Cheyenne,
Wyoming, filed an amicus curiae brief for the State of Wyoming.

M. Reed Hopper and Scott Sommerdorf, Pacific Legal Foundation, Sacramento,
California, filed an amicus curiae brief for Pacific Legal Foundation.

Before BRISCOE, EBEL, and HARTZ, Circuit Judges.

HARTZ, Circuit Judge.

The Center for Native Ecosystems, the Biodiversity Conservation Alliance,

and the Forest Guardians (collectively CNE) appeal the district court’s order

denying a petition for review of the United States Forest Service’s authorization

of livestock grazing in Medicine Bow National Forest. CNE first contends that

-3-
the Forest Service violated § 7(a)(2) of the Endangered Species Act, 16 U.S.C.

§ 1536(a)(2), because (1) its consultation with the United States Fish and Wildlife

Service (FWS) after the designation of portions of the forest as critical habitat for

the Preble’s meadow jumping mouse (Preble’s mouse) failed to consider how

grazing in the mouse’s critical habitat would affect its recovery, and (2) it must

reinitiate consultation with the FWS regarding the effects of grazing on the mouse

itself because grazing has exceeded previously established limits. CNE also

contends that the Forest Service has violated § 313(a) of the Clean Water Act

because it has not complied with Wyoming water-quality requirements “in the

same manner, and to the same extent as any nongovernmental entity,” 33 U.S.C.

§ 1323(a). The Pole Mountain Cattlemen’s Association, the Wyoming Stock

Growers Association, the Wyoming Farm Bureau Federation, and the Laramie

County Farmers Union (collectively the Cattlemen’s Association), along with the

Wyoming Association of Conservation Districts, intervened in the district-court

proceeding as defendants in support of the Forest Service’s actions. The State of

Wyoming, the Pacific Legal Foundation, and the National Association of Home

Builders in conjunction with the American Forest and Paper Association have

filed amicus briefs supporting various aspects of the Forest Service’s actions. We

have jurisdiction under 28 U.S.C. § 1291 and affirm the district court’s decision.

I. BACKGROUND

A. The Pole Mountain Area in Medicine Bow National Forest

-4-
 The Forest Service has long permitted livestock grazing in the Pole

Mountain area of Medicine Bow National Forest, near Laramie, Wyoming. Under

federal regulations the Forest Service may allow grazing on national forest land

by issuing an allotment management plan, 36 C.F.R. § 222.2, and grazing or

livestock-use permits, id. § 222.3(a). The allotment management plan must be

consistent with the land management plan for the area, id. § 222.2(c), which in

this case is the “Medicine Bow National Forest and Thunder Basin National

Grassland Land and Resource Management Plan” (the Forest Plan), issued in

October 1985.

The Pole Mountain allotment management plan allows grazing of up to

2086 cattle and 1200 sheep during an annual season from June 1 to October 15. It

divides Pole Mountain into eight livestock allotments, seven of which are used for

grazing. It also adopts certain best management practices for grazing, including a

prohibition on season-long grazing in a pasture, standards limiting the utilization

of forage by livestock, and the use of a deferred-rotation grazing system in which

“only one pasture in an allotment will be grazed at a time” and “the order in

which the pastures are used will be rotated each grazing season.” Aplts. App.

Vol. 2 at 409. Such practices are outlined in a publication of the Wyoming

Department of Environmental Quality entitled “Grazing Best Management

Practices.” Aplees. Jt. Supp. App. Vol. 2 at 378.

-5-
 Grazing permits, which generally are for a 10-year term, id. § 222.3(c)(1),

were issued for the seven Pole Mountain allotments in 1999. They identify the

maximum number of livestock and maximum length of grazing season for each

allotment. They also explain that they can

be cancelled, in whole or in part, or otherwise modified, at any time
during the [10-year] term to conform with needed changes brought
about by law, regulation, Executive order, allotment management
plans, land management planning, numbers permitted or seasons of
use necessary because of resource conditions, or the lands described
otherwise being unavailable for grazing.

Aplees. Jt. Supp. App. Vol. 3 at 575. The permits explicitly incorporate the

allotment management plan into their terms.

The limits set by the allotment management plan and permits on the length

of the grazing season and number of permissible livestock may be altered by

annual operating instructions issued by the Forest Service to grazing permittees.

Annual operating instructions are not required by any statute or regulation; but

the Forest Service Handbook for the Rocky Mountain Region contemplates their

use and describes their function: They specify the annual actions necessary to

implement the Forest Service’s decision to authorize grazing in a particular area.

They “identify the obligations of the permittee and the Forest Service, . . .

articulate annual grazing management requirements and standards, and [set forth

the] monitoring necessary to document compliance.” Aplts. App. Vol. 2 at 321.

They also take into account developments, such as a drought, occurring after

-6-
issuance of the allotment management plan and accordingly specify the maximum

amount of grazing authorized for a particular allotment, the precise sequence of

grazing on the allotment, and any other standards the permittee must follow that

year when grazing.

B. Facts Related to Claims Under the Endangered Species Act

In 1998 the FWS added the Preble’s mouse, which resided in areas of Pole

Mountain where grazing was authorized, to the threatened-species list. See

63 Fed. Reg. 26,517 (May 13, 1998). The FWS’s action triggered § 7(a)(2) of the

Endangered Species Act (ESA), 16 U.S.C. § 1536(a)(2), which generally requires

federal agencies—in this case the Forest Service—to consult with the FWS, on

behalf of the Secretary of the Interior, to “insure that any action authorized,

funded, or carried out by such agency . . . is not likely to jeopardize the continued

existence of any endangered species or threatened species or result in the

destruction or adverse modification of habitat of such species which is determined

. . . to be critical.” (For some species, federal agencies are required to consult

with the National Marine Fisheries Service on behalf of the Secretary of

Commerce, instead of the FWS. See Nat’l Ass’n of Home Builders v. Defenders

of Wildlife, Nos. 06-340 & 06-549, 2007 WL 1801745, at *5 (U.S. June 25,

2007); 50 C.F.R. § 402.01(b). That is not the case here.)

Following the threatened-species designation, the Forest Service began

preparing revisions to the Pole Mountain allotment management plan. As it

-7-
explained at the time, one of the reasons for doing so was to “[i]ncorporate

mitigation measures designed to protect sensitive and [threatened and endangered]

species into [the Pole Mountain allotment management plan].” Aplees. Jt. Supp.

App. Vol. 1 at 77. At the same time, the Forest Service sought to satisfy its

consultation obligation under 16 U.S.C. § 1536(a)(2). FWS regulations provide

that an agency’s consultation obligation may be satisfied through either formal or

informal consultation, depending on the agency’s determination of the possible

effect on the species or habitat at issue. If the agency action “may affect” the

species or habitat, 50 C.F.R. § 402.14(a), formal consultation is generally

required. If, however, “as a result of the preparation of a biological

assessment . . . or as a result of informal consultation with the [FWS], the . . .

agency determines, with the written concurrence of [the FWS], that the proposed

action is not likely to adversely affect” the species or habitat, id. § 402.14(b)(1),

formal consultation is not necessary, id.; §§ 402.14(a), 402.13(a).

In recognition of its ESA consultation obligation, the Forest Service

completed a biological assessment (the 1998 BA) analyzing the effects on the

Preble’s mouse of the proposed revisions to the allotment management plan. The

1998 BA identified nine “management requirements and mitigation measures”

designed to ensure the conservation of the mouse. Aplts. App. Vol. 2 at 393; see

id. at 392 (1998 BA) (“There currently exists adequate Forest Plan standards and

-8-
guidelines to allow for both livestock grazing in riparian areas and species

conservation.”). These measures were:

1. In areas where documented jumping mouse populations exist,
grazing management will maintain or enhance vegetative
habitat characteristics for the jumping mouse.
2. Prohibit season-long grazing in riparian pastures.
3. Implement winter grazing or short duration spring or late fall
grazing where possible to insure seed production for jumping
mouse forage during the majority of the grazing season.
4. Implement total rest in riparian pastures with deteriorated
range where conditions are not likely to improve with livestock
grazing.
5. Remove livestock from grazing units when average stubble
heights on carex species reach 3 to 4 inches in spring or winter
use pastures and 4 to 6 inches in summer/fall pastures.
6. Remove livestock from the grazing unit when streambank
disturbance (trampling, exposed soils, etc.), from current years
livestock grazing reaches 20 to 25 percent of the key area
stream reach.
7. Limit utilization of woody plants to 15 to 20 percent of current
animal growth.
8. Control the length of grazing period in spring use riparian
pastures to minimize utilization of re-growth. This is normally
20 to 30 days.
9. Limit utilization of herbaceous species to 40 to 45 percent.

Id. at 393–94. The 1998 BA also stated that “[l]ong-term trend monitoring shall

be conducted in representative riparian community types on a 3 to 5 year cycle to

determine effectiveness of the mitigation measures.” Id. at 394. It provided that

these measures would be added to the grazing permits. The 1998 BA concluded

that “[w]ith the implementation of the mitigation measures and monitoring,”

grazing was “not likely to adversely affect” the mouse or its habitat. Id. (internal

quotation marks omitted).

-9-
 On September 9, 1998, the FWS concurred with the 1998 BA’s conclusion

that the allotment-management-plan revisions, “as described, [are] not likely to

adversely affect Preble’s [mouse].” Id. at 384; see 50 C.F.R. § 402.14(b)(1)

(formal consultation not required “if, as a result of the preparation of a biological

assessment . . . or as a result of informal consultation with the [FWS], the . . .

agency determines, with the written concurrence of the [FWS], that the proposed

action is not likely to adversely affect any listed species or critical habitat”). On

October 19, 1998, the Forest Service informed the FWS that the appropriate

forage-utilization standard for the ninth mitigation measure listed in the 1998 BA

should be 45–55% rather than 40–45%; this change was necessary to ensure that

the 1998 BA was consistent with the 1985 Forest Plan, which provided that

“utilization on allotments,” Aplts. App. Vol. 2 at 421, was to be limited to

45–55% and that utilization levels could not exceed an amount 10% above the

limit. (This results in a total permissible utilization rate of 60.5%.) On

October 22, 1998, the FWS, having been advised of the proper forage-utilization

standard, once again “concur[red] with [the Forest Service’s] assessment that the

project, as described, is not likely to adversely affect Preble’s.” Id. at 373. The

same day, the Forest Service issued a Decision Notice and Finding of No

Significant Impact for the revisions to the Pole Mountain allotment management

plan. It added two mitigation measures to the nine that were previously identified

in the 1998 BA:

-10-
 10. Require the maintenance of a 4 inch stubble height of sedges
and rushes in all riparian areas within grazing allotments.
11. Prior to weed spraying or other vegetation management
activities (e.g. burning), site-specific analyses will be
conducted and Biological Evaluations will be prepared.

Id. at 366. It then concluded that the revisions “would not cause significant

environmental effects” and that no further environmental review was necessary.

Id. at 369. In particular, the revisions of the plan “‘may [a]ffect’ but [were] ‘not

likely to adversely affect’ the Preble’s meadow jumping mouse or its habitat.”

Id. at 371.

The following year, 1999, the Forest Service issued grazing permits for

each of the seven allotments; the 10-year permits incorporated the Pole Mountain

allotment management plan and established the maximum number of livestock

and season of use for each allotment. As explained above, they also provided that

the Forest Service may alter the grazing season and livestock numbers to meet the

objectives of the allotment management plan and the Forest Plan. Apparently

there were no administrative appeals after the permits were issued, and annual

operating instructions were later issued each year for each allotment.

In June 2003 the FWS designated certain areas of Pole Mountain as

“critical habitat” for the Preble’s mouse. 68 Fed. Reg. 37,276, 37,308, 37,321

(June 23, 2003). This critical habitat, the FWS’s notice explained, consisted of

those areas identified as essential to the mouse’s conservation. Id. at 37,295. The

notice defined conservation as “the use of all methods and procedures that are

-11-
necessary to bring [the mouse] to the point at which listing under the [Endangered

Species] Act is no longer necessary,” id., rather than merely measures ensuring its

survival and preventing extinction. The FWS incorporated into the mouse’s

critical-habitat designation and conservation strategy certain analysis from a

working draft of a recovery plan for the mouse, which “describe[d] actions

considered necessary for [its] conservation . . . , establish[ed] criteria for

downlisting or delisting the species, and estimate[d] time and cost for

implementing the recovery measures needed.” Id. at 37,280. The portion of the

Preble’s mouse’s critical habitat in Pole Mountain was “designated to address two

of three small recovery populations called for . . . in our conservation strategy.”

Id. at 37,308. It began along the eastern boundary of Pole Mountain and included

roughly 4.9 miles of streams in the North Pasture and Horse Creek allotments.

Critical habitat extended 360 feet from each side of these streams. Id. at 37,321.

In 2003 the Forest Service prepared a new biological assessment (the 2003

BA) analyzing the effects of a proposed revision to the Forest Plan on endangered

species, including the mouse, and their critical habitats in the area. It concluded

that “[t]here is no evidence of detrimental effects of livestock grazing on Preble’s

meadow jumping mouse, if the grazing meets Plan Standards.” Aplees. Jt. Supp.

App. Vol. 1 at 53. The 2003 BA added that “[t]he assumption that current

regulation of livestock grazing provides conditions compatible with the recovery

of the Preble’s meadow jumping mouse will be tested in a study of the effects of

-12-
grazing and fire on Preble’s meadow jumping mouse.” Id. Initially, it also found

that the proposed Plan’s provision for prescribed fires was likely to adversely

affect mice and their critical habitat. After formal consultation, however, the

FWS concluded that the planned prescribed fires were not likely to jeopardize the

continued existence of the mouse or adversely modify its critical habitat.

On March 16, 2004, CNE and others submitted to the Forest Service a

Notice of Intent to File Suit. They contended in part that “[s]ince the designation

of Critical Habitat [for the mouse], [the Forest Service] ha[d] not yet addressed

how [the] designation affect[ed] domestic livestock grazing in the Pole Mountain

Unit.” Aplts. App. Vol. 2 at 237. It added that “[b]y failing to prepare a new

biological assessment to address impacts to Preble’s meadow jumping mouse

Critical Habitat and by failing to reinitiate formal consultation with the [FWS],”

the Forest Service was violating the Endangered Species Act. Id. at 239.

Shortly thereafter the Forest Service met with the FWS to review the 1998

BA. As the Forest Service said at the time, its review considered the following

developments:

• Designation of Preble’s [meadow jumping mouse] critical
habitat within Pole Mountain Grazing Allotments.
• New livestock and grazing standards and guidelines in the
Revised Forest Plan[.]
• Results of forage utilization monitoring in Preble’s habitat[.]
• Ongoing drought conditions since 1998[.]

-13-
Aplts. App. Vol. 1 at 214. Among the review’s observations was that forage-

utilization levels specified in the 1998 BA had been exceeded in certain “key

areas [within allotments] where vegetation was specifically sampled.” Id. at 218.

(A later Forest Service report, apparently quoting a 1996 Forest Service

publication, defines key areas as “a portion of the range, which, because of its

location, grazing or browsing value, and/or use, serves as an indicative sample of

range conditions, trend, or degree of use seasonally. A key area guides the

general management of the entire area of which it is part.” Id. at 188 n.3 (internal

quotation marks omitted).) The review added, however, that these forage-

utilization measurements had been taken “in the hardest grazed areas of a pasture

with the intention that, if we protect these areas, the rest of the drainage is in

fairly good condition. As a result, exceedance of utilization standards in a key

area does not indicate that utilization was exceeded across the entire pasture or

riparian.” Id. at 218. The review noted that the 1998 BA had “not clarif[ied] if

utilization standards are to be met at a pasture level, on average across the

allotments, or within each key area measured,” and therefore recommended that a

supplement to the biological assessment should, among other things, “[i]dentify

the monitoring protocol used to evaluate [e]ffects to Preble’s mouse and their

habitat.” Id. at 221.

The informal consultation between the Forest Service and the FWS resulted

in a December 15, 2004, update to the 1998 BA. The update reviewed the nine

-14-
mitigation measures identified in the 1998 BA and the two mitigation measures

added in the 1998 Decision Notice and Finding of No Significant Impact, which

had found that the allotment-management-plan revisions would not cause

significant environmental effects. These 11 mitigation measures, according to the

update, had been “thought to be important to the conservation of” the Preble’s

mouse and its habitat. Id. at 136. The update analyzed forage utilization by

considering the average rate of utilization for all key areas within a given

allotment. It explained:

By design, utilization cages were established in the most heavily
grazed areas of a pasture with the intention that, if we protect these
areas, the rest of the drainage would be in fairly good condition. As
a result, higher utilization in a key area does not indicate that
utilization was exceeded across the entire riparian area, pasture or
allotment.

Id. at 143. The update concluded that forage-utilization standards were being met

as of 2004. Average utilization exceeded 55% in only one allotment, North

Pasture, where it was 59%, still within the permissible 60.5% limit. Furthermore,

the only three key areas in Pole Mountain within the Preble’s mouse’s critical

habitat all had permissible rates of utilization. The update observed that the 11

mitigation measures were accomplished as of 2004, “leading to a stable or

improving trend in riparian areas. These areas provide the mainstay of Preble’s

habitat and are most important to conservation of the species.” Id. at 146. In

light of this analysis, the Forest Service determined that “[t]he effects occurring

-15-
to Preble’s mouse from current grazing practices are the same as those considered

in the [1998 BA].” Id. at 147.

Consistent with the December 15 update, on December 30, 2004, the Forest

Service concluded that grazing in Pole Mountain was “[n]ot [l]ikely to [a]dversely

[a]ffect [c]ritical [habitat] for the Preble’s mouse.” Id. at 124 (internal quotation

marks omitted). It made no determination, however, whether grazing would

adversely affect the mouse, an issue that had previously been addressed in the

1998 BA. In reaching its conclusion regarding the mouse’s critical habitat, the

Forest Service summarized the information in the December 15 update, observing

in particular that “utilization in the Horse Creek Allotment and the North Pasture

Allotment (which contain critical habitat) was within 1985 Forest Plan standards.”

Id. at 126. On January 12, 2005, the FWS concurred with the Forest Service’s

conclusion regarding the effects of grazing on critical habitat. It based its

concurrence on information from the Forest Service, including that there was

“appropriate utilization (meeting the 1985 Forest Plan standards) within the

allotments.” Id. at 121.

C. Facts Related to Claim Under the Clean Water Act

In 2000 the Pole Mountain area began suffering from a drought.

Consequently, by 2002 the Forest Service was advising grazing permittees of the

need to make operational adjustments. The 2002 annual operating instructions

(AOIs) instituted moderate reductions in the amount of authorized grazing.

-16-
Through a combination of the AOIs’ reductions and apparently voluntary

adjustments by permittees, cattle grazing in Pole Mountain was reduced by 48%

from the maximum allowed under the permits.

In October 2002 the Wyoming Department of Environmental Quality

(WDEQ) found that at one of three tested locations in Pole Mountain—North

Branch North Fork Crow Creek, in the Crow Creek allotment—the state water-

quality standard for fecal coliform was exceeded. See 020-080-001 Wyo. Code R.

§ 27 (Weil 2007) (fecal-coliform standard). The level of coliform bacteria is used

as an indicator of possible sewage contamination because they are commonly

found in human and animal feces and suggest the presence of pathogenic bacteria,

viruses, and protozoans. See Envtl. Prot. Agency, What Are Fecal Bacteria and

Why Are They Important?, available at http://www.epa.gov/volunteer/stream/

vms511.html. The Forest Service was notified of the WDEQ’s finding in

November 2002.

For the 2003 grazing season, AOIs for Pole Mountain allotments further

reduced the amount of authorized grazing. In Crow Creek, where the high fecal-

coliform readings had been taken, the 2003 AOIs authorized only 1559 animal

months of grazing, significantly less than the 2047 animal months allowed under

the grazing permits and the 1932 animal months allowed under the 2002 AOIs.

Ultimately there were only 1253 actual animal months of grazing in Crow Creek

in 2003, a 39% reduction from the amount allowed under the grazing permits.

-17-
Overall, 2003 cattle grazing in Pole Mountain was 47% lower than that allowed

under the permits.

The WDEQ took more samples in 2003. Samples in the spring showed no

excessive levels of fecal coliform; fall samples, however, revealed excessive

levels once again at North Branch North Fork Crow Creek in the Crow Creek

allotment and now also at Middle Crow Creek in the Green Mountain allotment.

As a result, in 2004 the State of Wyoming added these stream areas to its Clean

Water Act (CWA) list of waters not meeting state standards. See 33 U.S.C.

§ 1313(d)(1); 40 C.F.R. § 130.7. The CWA requires states to prioritize all waters

not meeting state standards, “taking into account the severity of the pollution and

the uses to be made of such waters.” 33 U.S.C. § 1313(d)(1)(A). The

prioritization identifies those waters for which the state will first calculate the

maximum daily load of pollutants that the body of water can accept without

violating water-quality standards. See id. § (d)(1)(C); 40 C.F.R. § 130.2(I)

(definition of total maximum daily load); id. § (f) (water’s loading capacity is

greatest amount of pollutant it can receive without violating water-quality

standards). This process can take several years. See id. § 130.7(b)(4) (“The

priority ranking shall specifically include the identification of waters targeted for

[total maximum daily load] development in the next two years.”). The stream

areas in Pole Mountain were given a low priority for development of these

standards. WDEQ justified this low prioritization on the ground that the water

-18-
quality in these areas was to be addressed by the Crow Creek Watershed Steering

Committee, which was comprised of conservation districts, local-government

officials, and various interest groups.

Also in 2004 the Forest Service prepared a Water Quality Action Plan

identifying the measures that it would take “to ensure [that] water quality in the

impaired stream segments of the North Branch North Fork Crow Creek and

Middle Crow Creek will consistently meet Wyoming State DEQ standards.”

Aplees. Jt. Supp. App. Vol. 1 at 239. It once again reduced stock numbers from

the maximums allowed under the permits by limiting authorization for grazing in

the AOIs for the Crow Creek and Green Mountain allotments. The 2004 AOIs for

Crow Creek authorized only 1368 animal months of grazing compared to the 2047

specified in the grazing permits and the 1559 authorized in the 2003 AOIs. The

2004 AOIs for Green Mountain authorized 1710 animal months of grazing

compared to the 2252 specified in the grazing permits and 1535 in the 2003 AOIs.

Ultimately, grazing in the Crow Creek allotment was 43% lower than that allowed

under the permits, and grazing in the Green Mountain allotment was 30% lower.

At the end of the 2004 grazing season, only one location, North Branch

North Fork Crow Creek, exceeded fecal-coliform standards, although the level

was less than one-fourth of that in 2003. The Forest Service and conservation

districts entered into a Memorandum of Understanding to amend the Crow Creek

Watershed plan to address the high bacteria levels in the area. As the

-19-
memorandum explains, “The watershed plan would identify potential pollution

sources and identify practices . . . intended to improve water quality.” Id. Vol. 2

at 257.

D. District-Court Proceedings

On November 19, 2004, CNE filed a complaint in the United States District

Court for the District of Colorado, naming the Forest Service and Rick Cables,

Regional Forester for Region 2, as defendants. On December 15, 2004, the Forest

Service submitted to the FWS its update to the 1998 BA. In an amended

complaint, also filed on December 15, and a later-filed petition for review, CNE

challenged the Forest Service’s actions under two statutes. First, it claimed that

the Forest Service violated § 7(a)(2) of the Endangered Species Act, 16 U.S.C.

§ 1536(a)(2), because (1) its consultation on critical habitat did not consider the

effect of grazing on the mouse’s recovery, but only on its survival; and (2) it had

not reinitiated consultation regarding grazing’s effect on the mouse itself even

though there had been violations in multiple key areas of the forage-utilization

standards in the 1998 BA. Second, it claimed that the Forest Service violated

§ 313(a) of the CWA, 33 U.S.C. § 1323(a), because fecal-coliform levels in the

Pole Mountain area exceeded Wyoming standards. The Cattlemen’s Association

and the Wyoming Association of Conservation Districts intervened as defendants.

The Forest Service and the intervenors moved to dismiss the complaint on

-20-
multiple grounds, including lack of final agency action. The district court denied

the motions on September 29, 2005.

The district court denied CNE’s petition for review of agency action on

January 9, 2006. On the claims under the Endangered Species Act, the court

ruled that the Forest Service’s consultations were reasonable. As to the claim

under the CWA, it said that the Forest Service “appears to be complying with

state water quality standards to the same extent as any nongovernmental entity”

and accordingly concluded that 33 U.S.C. § 1323(a) was satisfied. Aplts. App.

Vol. 1 at 32 (Order on Pet. for Review, Jan. 9, 2006) (internal quotation marks

omitted). Final judgment was entered on February 14, 2006.

II. DISCUSSION

A. Standard of Review

CNE’s claims in this case are governed by the Administrative Procedure

Act (APA). 5 U.S.C. § 706; see Bennett v. Spear, 520 U.S. 154, 174–75, 179

(1997) (claim under 16 U.S.C. § 1536 reviewed under APA); Ore. Natural Res.

Council v. U.S. Forest Serv., 834 F.2d 842, 851 (9th Cir. 1987) (judicial review of

33 U.S.C. § 1323 claim under APA is appropriate). Although “[o]ur standard of

review of the lower court’s decision in an APA case is de novo,” N.M. Cattle

Growers Ass’n v. U.S. Fish & Wildlife Serv., 248 F.3d 1277, 1281 (10th Cir.

2001), the APA narrows the scope of our review of an agency’s actions, see id.

Under the APA we set aside the agency’s action only if it is “arbitrary,

-21-
capricious, an abuse of discretion, or otherwise not in accordance with law.”

§ 706(2)(A).

The duty of a court reviewing agency action under the “arbitrary and
capricious” standard is to ascertain whether the agency examined the
relevant data and articulated a rational connection between the facts
found and the decision made. In reviewing the agency’s explanation,
the reviewing court must determine whether the agency considered
all relevant factors and whether there has been a clear error of
judgment.

Olenhouse v. Commodity Credit Corp., 42 F.3d 1560, 1574 (10th Cir. 1994)

(footnote and citation omitted). We review each of CNE’s claims in turn.

B. Endangered Species Act Claim

CNE challenges the adequacy of the Forest Service’s consultation with the

FWS after the Preble’s mouse’s critical habitat was designated in 2003. CNE first

contends that the Forest Service violated the Endangered Species Act (ESA)

because its consultation with the FWS considered only how livestock grazing in

the mouse’s critical habitat would affect its survival or preservation and not how

the grazing would affect its recovery. CNE also contends that the Forest Service

failed to reinitiate consultation on the effects of grazing on the mouse itself, even

though the forage-utilization standards identified in the 1998 BA had been

violated in multiple key areas. We reject both contentions.

1. Recovery

CNE argues that the Forest Service’s consultation after designation of

critical habitat violated the ESA because it “did not consider how livestock

-22-
grazing in critical habitat would impact the mouse’s recovery and therefore did

not satisfy the basic legal requirement of section 7 of the ESA.” Aplts. Br. at 18.

It asserts that agencies must insure that actions not only prevent the extinction of

species but also allow for the recovery of the species, that is, allow the species to

increase sufficiently in population that it can be removed from the list of

endangered or threatened species (an action referred to as “delisting,” see 50

C.F.R. § 424.11(d)).

We agree with CNE’s interpretation of the law governing this case. Section

7(a)(2) of the ESA, 16 U.S.C. § 1536(a)(2), describes a federal agency’s duty to

consult:

Each Federal agency shall, in consultation with and with the
assistance of the Secretary [of the Interior, see § 1532(15)], insure
that any action authorized, funded, or carried out by such agency . . .
is not likely to jeopardize the continued existence of any endangered
species or threatened species or result in the destruction or adverse
modification of habitat of such species which is determined by the
Secretary [of the Interior], after consultation as appropriate with
affected States, to be critical . . . .

The Forest Service is thus obligated to insure, through consultation with the FWS

on behalf of the Secretary of the Interior, that its action “is not likely to . . . result

in the destruction or adverse modification of” the Preble’s mouse’s critical

habitat. FWS regulations explain that it may satisfy this obligation through either

formal or informal consultation. As a general matter, formal consultation is

required when agency action “may affect listed species or critical habitat.”

-23-
50 C.F.R. § 402.14(a). An agency may forgo formal consultation, however, if it

engages in informal consultation with the FWS and determines, with the written

concurrence of the FWS, that even if the proposed action “may affect listed

species or critical habitat,” id., it “is not likely to adversely affect any listed

species or critical habitat,” id. § (b)(1). Informal consultation “includes all

discussions, correspondence, etc., between the [FWS] and the Federal agency.”

Id. § 402.13(a).

Although neither § 1536(a)(2) nor the consultation regulations expressly

describe what must be considered by a federal agency during consultation on

critical habitat, the ESA’s definitions clarify the matter. In relevant part the

definition of critical habitat is “the specific areas within the geographical area

occupied by the species, at the time it is listed . . . , on which are found those

physical or biological features (I) essential to the conservation of the species and

(II) which may require special management considerations or protection.”

16 U.S.C. § 1532(5)(A)(I) (emphasis added). Thus, critical habitat is impaired

when features essential to the species’ conservation are impaired. The definition

of conservation is found in § 1532(3), which states that it “mean[s] to use and the

use of all methods and procedures which are necessary to bring any endangered

species or threatened species to the point at which the measures provided pursuant

to this chapter are no longer necessary.” Under this definition, conservation

encompasses recovery. See also 68 Fed. Reg. at 37,280 (incorporating

-24-
information from Preble’s mouse’s recovery plan, which describes the actions

“necessary for conservation of the species,” into designation of critical habitat).

It follows that critical habitat is “adverse[ly] modif[ied]” by actions that

adversely affect a species’ recovery and the ultimate goal of delisting. 1

Accordingly, we agree with CNE’s view of what the Forest Service was

required to do: Section 1536(a)(2) requires federal agencies, when considering

the effect of their actions on a species’ critical habitat, to consider the effect of

those actions on the species’ recovery. Contrary to CNE’s contention, however,

we read the record as showing that the Forest Service did what was required. As

the Forest Service points out on appeal, after the critical habitat for the Preble’s

1
We note that the FWS has promulgated a regulation to define when an
agency’s action results in “destruction or adverse modification.” 16 U.S.C.
§ 1536(a)(2). The regulation defines destruction or adverse modification as
occurring only when an action “appreciably diminishes the value of critical
habitat for both the survival and recovery of a listed species.” 50 C.F.R. § 402.02
(emphasis added). This definition was rejected in Gifford Pinchot Task Force v.
U.S. Fish & Wildlife Service, 378 F.3d 1059, 1069–71 (9th Cir. 2004), and Sierra
Club v. U.S. Fish & Wildlife Service, 245 F.3d 434, 441–43 (5th Cir. 2001), on
the ground that by requiring an effect on both the survival and recovery of a
species, the regulation “reads the ‘recovery’ goal out of the adverse modification
inquiry; a proposed action ‘adversely modifies’ critical habitat if, and only if, the
value of the critical habitat for survival is appreciably diminished,” Gifford
Pinchot, 378 F.3d at 1069. We have previously recognized the questionable
validity of this definition. See N.M. Cattle Growers, 248 F.3d at 1283 n.2
(“[F]ederal courts have begun to recognize that the results [that the regulatory
definition of adverse modification] produce[s] are inconsistent with the intent and
language of the ESA.”). But the Forest Service does not rely on this regulatory
definition to support its actions, and on December 9, 2004, the FWS apparently
instructed its biologists not to rely on the definition pending adoption of a new
definition. Therefore, we need not consider the validity of the definition in
§ 402.02.

-25-
mouse was designated in June 2003, it did in fact consider recovery when it

considered the effect of grazing on the conservation of the Preble’s mouse.

In 2004, shortly after CNE’s notice to file suit was submitted, the Forest

Service met with the FWS to review the 1998 BA because of several interim

developments, including “[d]esignation of Preble’s [mouse] critical habitat within

Pole Mountain Grazing Allotments.” Aplts. App. Vol. 1 at 214. The area of Pole

Mountain identified as the Preble’s mouse’s critical habitat in 2003 had been

expressly “designated to address two of three small recovery populations called

for . . . in [the FWS’s] conservation strategy.” 68 Fed. Reg. at 37,308 (emphasis

added). As the FWS’s official notice of designation explained: “Restoring an

endangered or threatened species to the point where it is recovered is a primary

goal of our endangered species program.” Id. at 37,280. In light of the purpose

of the critical-habitat designation—recovery of the mouse—it is hard to see how

the Forest Service’s review in 2004, and the FWS’s ultimate concurrence with the

conclusions of that review, could have been directed at anything but recovery.

In any event, the review undeniably considered recovery by considering

conservation. On December 30, 2004, the Forest Service sent to the FWS for

concurrence the report by its wildlife biologist determining that “livestock

grazing, as described in the Pole Mountain AMPs and implemented annually, is

‘Not Likely to Adversely affect Critical Habitat for the Preble’s mouse.’” Aplts.

App. at 124. That report relied on prior reports, including the Forest Service’s

-26-
December 15, 2004, update to the 1998 BA. The update, which reviewed the 11

previously identified mitigation measures, stated that measures considered in the

1998 BA and again in the update were “important to the conservation of Preble’s

meadow jumping mice and their habitat.” Aplts. App. Vol. 1 at 136. Most

importantly, the update’s conclusions addressed conservation. After a detailed

review of the 11 measures, the update concluded that they were satisfied, “leading

to a stable or improving trend in riparian areas. These areas provide the mainstay

of Preble’s habitat and are most important to conservation of the species.” Id. at

146 (emphasis added). And the update’s plan of action was also directed at

conservation. To counteract recent indications of “more intense grazing pressure

in some areas of Preble’s mouse habitat,” said the update, the Forest Service was

taking additional actions, including “reduced stocking, increased herding

[apparently referring to movement of livestock], and pasture rotation adjustments

in order to continue to provide the necessary conservation measures for Preble’s

mouse.” Id. at 147 (emphasis added). Accordingly, it is clear that (1) the Forest

Service had considered the conservation of Preble’s mouse when it forwarded the

update to the FWS on December 30, 2004, and stated its determination that

“livestock grazing . . . is Not Likely to Adversely Affect Critical Habitat for the

Preble’s mouse,” id. at 124 (internal quotation marks omitted), and (2) the FWS

likewise had considered conservation when it concurred in writing on January 12,

2005.

-27-
 CNE nevertheless claims that “‘implicit[]’” consideration of recovery

violates “basic APA judicial review principles.” Aplts. Reply Br. at 3. We

assume that CNE’s argument is targeted at the Forest Service’s alleged failure to

use the word recovery in the 2004 update to the 1998 BA. But, as noted above,

the ESA itself speaks only in terms of conservation, not recovery. It states that

critical habitat is habitat “essential to the conservation of the species,” 16 U.S.C.

§ 1532(5)(A)(I) (emphasis added), and that conservation is the methods and

procedures which are necessary to allow the species to recover “to the point at

which the measures provided pursuant to [the ESA] are no longer necessary,” id.

§ 1532 (3). We cannot fault the Forest Service for employing the statutory term.

Criticism would be more justifiable if it did not. For this reason, CNE’s reliance

on Gifford Pinchot, 378 F.3d at 1072 n.9, is unavailing. In that case the Ninth

Circuit rejected the Forest Service’s position because it had failed to consider

explicitly either recovery or conservation. See id.; id. at 1072–73. That is not a

problem here.

CNE also insists that the Forest Service could not have considered recovery

because its 2004 consultation on critical habitat addressed no issues that it had

not considered in 1998 when it consulted on the species itself. But the questions

to be answered on the two occasions are intimately related, so it would not be

surprising that the same considerations would control the answers. We find it

significant that CNE never identifies for us any particular issue or factor relevant

-28-
to recovery of the mouse that the Forest Service failed to consider in 2004. In

short, the Forest Service’s consideration of the effect of grazing on the

conservation of the Preble’s mouse complied with 16 U.S.C. § 1536(a)(2). See

5 U.S.C. § 706(2)(A).

CNE half-heartedly argues that the Forest Service did not adequately raise

below the ground on which we rely to affirm the district court’s ruling. In a

footnote in its brief-in-chief, and without citation to the record, it asserts that

“The [Forest Service] did not argue to the district court that it did address

recovery in this process, but rather that it need not consider recovery.” Aplts. Br.

at 24 n.5. But even if the assertion is correct, we may affirm a district-court

judgment on any ground appearing from the record so long as the litigants had a

fair opportunity to develop the record, see Garrison v. Gambro, Inc., 428 F.3d

933, 939 (10th Cir. 2005), and to address the ground on which we rely, see Gomes

v. Wood, 451 F.3d 1122, 1133 (10th Cir. 2006). The purpose of requiring

presentation of the issue in the lower court is “to ensure that litigants may not be

surprised on appeal by final decision there of issues upon which they have had no

opportunity to introduce evidence or to present whatever legal arguments they

may have.” Anixter v. Home-Stake Prod., 77 F.3d 1215, 1228 (10th Cir. 1996)

(ellipsis, brackets, and internal quotation marks omitted). There is no unfairness

here. First, because judicial review is based on the administrative record, see

5 U.S.C. § 706, we doubt that CNE could have introduced further evidence on the

-29-
matter, and it does not suggest otherwise. And second, CNE had ample

opportunity to present its legal arguments in its briefs and at oral argument to this

court. Indeed, its appellate opening brief argues that the Forest Service had not

considered recovery.

In sum, we conclude that the Forest Service’s analysis of the effect of its

actions on the conservation of the mouse, and the FWS’s concurrence with that

analysis, satisfied § 1536(a)(2)’s requirement that recovery be considered.

2. Forage Utilization 2

CNE next contends that the Forest Service has violated the 1998 BA’s

45–55% forage-utilization standard for Pole Mountain and that the violations

require reinitiation of consultation to consider the effects of the allegedly

excessive forage utilization on Preble’s mouse. See 50 C.F.R. § 402.16 (defining

circumstances in which reinitiation of consultation is required). Although in

2004–2005 the Forest Service informally consulted with the FWS regarding the

effects of grazing on the mouse’s critical habitat, the FWS’s concurrence at the

end of this consultation did not explicitly address the effects of grazing on the

mouse itself, a topic that had originally been addressed in the 1998 BA. CNE

claims that the Forest Service must reinitiate consultation regarding the effects of

grazing on the mouse either because “new information reveal[ed] effects of the

2
We express no view on the merits of the ground relied on by Judge
Briscoe’s concurrence in disposing of this issue.

-30-
action that may affect listed species . . . in a manner or to an extent not previously

considered,” id. § (b), or because “the identified action [was] subsequently

modified in a manner that cause[d] an effect to the listed species . . . that was not

considered in the biological opinion,” id. § (c).

We do not agree. Paragraphs (b) and (c) of § 402.16 both require

reinitiation of consultation only when the effects to species that are revealed or

caused are different from those effects previously considered. See 51 Fed. Reg.

19,926, 19,956 (June 3, 1986) (“[50 C.F.R. § 402.16(c)] show[s] that changes to

the action that do not cause effects different from or additional to those

considered in the biological opinion will not require reinitiation of formal

consultation.”). But here the Forest Service found just the opposite. In its

December 15, 2004, update to the 1998 BA, it reviewed the mitigation measures

designed to minimize the effect of grazing on Preble’s mouse. In doing so it

measured forage utilization in each allotment in Pole Mountain by averaging the

utilization rates for the key areas within the allotment. Under this approach it

found that the maximum forage-utilization rate of 60.5% was not exceeded in any

allotment, and that, as a result, “[t]he effects occurring to Preble’s mouse from

current grazing practices are the same as those considered in the [1998 BA].”

Aplts. App. Vol. 1 at 147 (emphasis added). Its decision not to reinitiate

consultation, see § 402.16(b), (c), thus did not violate the ESA.

-31-
 The Ninth Circuit’s decision in Forest Guardians v. Johanns, 450 F.3d 455

(9th Cir. 2006), is not contrary to our conclusion. In Forest Guardians the court

held that “[t]he material inadequacy of the Forest Service’s utilization monitoring

and the results of the limited measurements that were taken constituted

modifications to the allotment’s land management plan that affected listed species

in a manner and to an extent not previously considered.” Id. at 465. It explained

its holding as follows:

We do not hold that each isolated instance in which the Forest
Service deviated from [the allotment’s] guidance criteria [on which
the FWS’s concurrence in the “not likely to adversely affect” finding
was premised] required the agency to re-initiate consultation. The
Forest Service’s arguments to the contrary notwithstanding, the case
before us is not comprised of infrequent and insignificant deviations.
Rather, the undisputed facts are that (1) the guidance criteria
expressly stated that the utilization levels specified by the land
management plan were necessary to protect the ESA-listed species in
[a particular grazing allotment], (2) the Forest Service regularly
failed to meet the monitoring requirements on which the “not likely
to adversely affect” determination for those species was premised,
and (3) the evidence that the Forest Service did obtain as a result of
its deficient monitoring suggested that maximum permissible
utilization levels were being exceeded. In light of these facts, the
Forest Service’s failure to re-initiate consultation violated the ESA.

Id. at 465–66. We agree that reinitiation of consultation would be required if (1)

the FWS’s concurrence in a “not likely to adversely affect” finding expressly

required utilization levels to be met in order for the concurrence to remain valid,

(2) utilization levels were not monitored as specified by the FWS, and (3) the

monitoring that was conducted showed excess utilization. In that event, the

-32-
Forest Service could not properly assert that the effects to species were not

different from those previously considered. But that is not the circumstance here.

As we shall show, neither the second nor the third condition has been met: The

Forest Service has not failed to monitor utilization in Pole Mountain, and

utilization rates have not exceeded the 1998 BA’s standards. (We express no

opinion on whether compliance with the utilization rates was essential to the

FWS’s concurrence. And we need not decide whether reinitiation of consultation

would be required if just conditions (1) and (2) or conditions (1) and (3) were

satisfied.)

We first address utilization rates. CNE acknowledges that the Forest

Service determined in 2004 that forage-utilization standards were being met (so

that the effects of grazing on the mouse were the same as those considered in the

1998 BA). It claims, however, that this determination was the result of an

arbitrary change in the way that the Forest Service analyzed forage utilization. It

contends that after its complaint was filed the Forest Service shifted from

considering utilization in each key area separately to averaging utilization in all

key areas in an allotment, and that this change allowed the Forest Service to avoid

finding that forage utilization in Pole Mountain in 2004 exceeded standards. We

are not persuaded.

To begin with, the mitigation measures in the 1998 BA do not refer

specifically to key areas. The 1998 BA specified only that the Forest Service was

-33-
to “[l]imit utilization of herbaceous species to 40 to 45 percent.” Aplts. App.

Vol. 2 at 394. (Later that year, to ensure that the mitigation measures were

consistent with the 1985 Forest Plan, the Forest Service, with the FWS’s

concurrence, changed the permissible level of forage utilization to 45–55%. The

Forest Plan had previously specified that “utilization on allotments” was not to

exceed an amount 10% greater than the permissible utilization rate, id. at 421,

thus setting the actual upper limit at 60.5%.) Although CNE is correct that the

Forest Service uses key areas to measure forage utilization, the 1998 BA did not

give key-area utilization rates independent significance. Rather, key-area data,

according to a 1996 Forest Service publication, “serves as an indicative sample of

range conditions, trend, or degree of use seasonally. A key area guides the

general management of the entire area of which it is part . . . .” Id. Vol. 1 at 188

n.3 (internal quotation marks omitted). The Forest Service recognized that

excessive utilization in one key area (of several in an allotment) was not

representative of range conditions across the entire allotment. The December

2004 update to the 1998 BA explained:

By design, utilization cages were established in the most heavily
grazed areas of a pasture with the intention that, if we protect these
areas, the rest of the drainage would be in fairly good condition. As
a result, higher utilization in a key area does not indicate that
utilization was exceeded across the entire riparian area, pasture or
allotment.

...

-34-
 Localized areas of high grass utilization doesn’t necessarily mean
that there is a loss or decline of riparian condition. This lack of
connection is shown by the facts that despite localized areas of high
grass utilization, Forest Service monitoring does not show an overall
change in vegetation type or a downward trend in stream stability.
The monitoring does show an improving trend in shrub density,
vigor, and recruitment.

Id. at 143–44; see id. at 146 (in the Pole Mountain area, “other indicators of

riparian health ([heights of] shrubs and carex) demonstrate a stable or improving

trend in habitat”). Consequently, the Forest Service measured utilization by

averaging forage-utilization rates for all key areas within a particular allotment,

on the ground that this would be more accurate.

As for CNE’s contention that such averaging had not been used before

2004, there is evidence to the contrary, and, in any event, the Forest Service was

not bound to continue its pre-2004 practice. The 1985 Forest Plan explained that

the focus of monitoring was to determine “utilization [of forage] on allotments,”

id. Vol. 2 at 421 (emphasis added), and said nothing about key areas.

Furthermore, the Forest Service’s 1998 communication with the FWS, in which it

sought concurrence with its “not likely to adversely affect” finding, explained that

the forage-utilization standard helped guide how it “manage[d] the allotments on

Pole Mountain,” id. at 374 (emphasis added). If it is utilization on the allotment

that is important, one can infer that the proper measure is the average utilization

throughout the allotment, from which it inevitably follows that the utilization on

some portions will exceed the average and on some will fall below. Such

-35-
averaging can be found in a 1998 environmental assessment of the revisions to the

Pole Mountain allotment management plan, which lists utilization rates in each

key area and then states the average in each allotment. Thus, the Forest Service

apparently had used such averages for several years. And nothing within the

1998 BA suggests that utilization needed to be evaluated separately for each key

area.

On the other hand, we recognize that there is some doubt concerning

whether averaging was to be used in assessing forage utilization. As noted in a

2004 report supplementing a 1998 environmental assessment of the revisions to

the Pole Mountain allotment management plan, “The amended 1998 BA and

consultation do not clarify if utilization standards and guidelines are to be met at

a pasture level, on average across the allotments, or within each key area

measured.” Id. Vol. 1 at 209. Thus, the use of averaging may have constituted a

change in methodology. But even if averaging was initiated in 2004, change is

not forbidden. An agency is not bound by its prior position. “The law does not

require an agency to stand by its initial policy decisions in all circumstances.”

Exxon Corp. v. Lujan, 970 F.2d 757, 762 n.4 (10th Cir. 1992). Changes in policy

can be upheld when such change is explained with a reasoned analysis. See id.

And in evaluating whether the analysis is reasoned, we must defer to the agency’s

expertise. See Wyoming v. United States, 279 F.3d 1214, 1240 (10th Cir. 2002)

(“[D]eference to agency action is appropriate where that action implicates

-36-
scientific and technical judgments within the scope of agency expertise.” (internal

quotation marks omitted)); Custer County Action Ass’n v. Garvey, 256 F.3d 1024,

1036 (10th Cir. 2001) (“[T]he agency, not a reviewing court, is entrusted with the

responsibility of considering the various modes of scientific evaluation and theory

and choosing the one appropriate for the given circumstances.” (brackets and

internal quotation marks omitted)). As the above discussion demonstrates, the

Forest Service has provided a reasoned basis for concluding that a single key-area

utilization rate may be misleading and that its averaging methodology is the

proper measure of forage utilization in an allotment.

Moreover, the FWS accepted the validity of the Forest Service’s averaging

methodology in January 2005 when it concurred in the “not likely to adversely

effect” finding for the Preble’s mouse’s critical habitat. The FWS’s concurrence

at that time was expressly based on the Forest Service’s finding that there was

“appropriate utilization (meeting the 1985 Forest Plan standards [that set a limit

of 60.5%]) within the allotments.” Aplts. App. Vol. 1 at 121. The Forest

Service’s finding, in turn, relied on the averaging of key areas within each

allotment. Thus, the FWS itself relied on the average rate of forage utilization

within an allotment’s key areas to analyze the effects of grazing. The FWS’s

acceptance of the validity of this methodology for analyzing effects under the

ESA supports the conclusion that the methodology is hardly arbitrary. (Perhaps it

-37-
is also worth observing that even without averaging, the utilization standard was

not exceeded in 2004 in any of the three key areas within the critical habitat.)

A second premise of Forest Guardians—that the Forest Service’s

monitoring of forage utilization had been deficient, 450 F.3d at 466—is also not

present in this case. CNE claims that the Forest Service’s collection of forage-

utilization data was inadequate because it was based on “[o]cular [e]stimate[s].”

Aplts. App. Vol. 1 at 155. Citing only to some 2004 annual operating instructions

indicating that “[u]tilization will be determined using the clipped plant weight

method,” id. at 175, and a 2004 report noting that in previous years the Forest

Service had used ocular estimates in “low budget years,” id. at 192, CNE asserts

that “[q]uantitative monitoring data is mandated by the agency’s own

requirements.” Aplts. Br. at 30. The record is to the contrary. The 1985 Forest

Plan explains that “[t]hree methods may be used to determine utilization on

allotments.” Aplts. App. Vol. 2 at 421. One of these three “standard procedures”

is “visual estimates of grazing use.” Id. Furthermore, the same 2004 report cited

by CNE explains that ocular estimates were the customary means of monitoring

unless they revealed excessive utilization. As the report explains, “If ocular

estimates indicated that a key area was not exceeding [utilization] standards,

additional data were not collected.” Id. Vol. 1 at 193.

The Forest Service’s 2004 conclusion that the effects of grazing were the

same as those considered in 1998 was not arbitrary or capricious, and hence its

-38-
determination that it need not reinitiate consultation on the effects of grazing on

the Preble’s mouse did not violate the ESA.

C. Clean Water Act

CNE next contends that the Forest Service’s issuance of annual operating

instructions (AOIs) in 2003 and 2004 for grazing in the Crow Creek and Green

Mountain allotments was arbitrary and capricious under the APA and violated the

CWA because the Forest Service did not protect water quality in those allotments

to the same extent as required of private parties under CWA § 313(a), 33 U.S.C.

§ 1323(a). Section 1323(a) provides in pertinent part:

Each department, agency, or instrumentality of the executive,
legislative, and judicial branches of the Federal Government . . .
engaged in any activity resulting, or which may result, in the
discharge or runoff of pollutants . . . shall be subject to, and comply
with, all Federal, State, interstate, and local requirements,
administrative authority, and process and sanctions respecting the
control and abatement of water pollution in the same manner, and to
the same extent as any nongovernmental entity . . . .

Id.

We hold that issuance of the AOIs was lawful. Before our discussion of the

merits, however, we must first address a challenge to the district court’s

jurisdiction to consider CNE’s challenge.

1. Jurisdiction

The APA, under which CNE’s CWA claim is brought, see Ore. Natural

Res. Council v. U.S. Forest Serv., 834 F.2d 842, 851 (9th Cir. 1987), limits

-39-
judicial review not otherwise provided by statute to “final agency action for

which there is no other adequate remedy in a court.” 5 U.S.C. § 704. In the

district court CNE identified the Forest Service’s issuance of AOIs for Crow

Creek and Green Mountain as the basis for its CWA claim. The Forest Service

and the Wyoming Association of Conservation Districts contend that the Crow

Creek and Green Mountain AOIs issued in 2003 and 2004 for grazing in Pole

Mountain do not constitute final agency action. As the Forest Service argues,

“[i]t is the [grazing] permits that grant permission to graze livestock on the

allotments,” Aplees. (Service) Br. at 48, whereas AOIs “are merely a tool for

implementing the decisions made in the [allotment management plan] and

permits,” id. at 49. We disagree and hold that the AOIs are final agency action.

The APA defines agency action to include “the whole or a part of an

agency . . . license.” 5 U.S.C. § 551(13). License is defined to include “the

whole or a part of an agency permit.” Id. § (8). The Crow Creek and Green

Mountain AOIs are licenses because, as they expressly state, they are “included as

part of” the previously issued grazing permits. E.g., Aplts. App. Vol. 1 at 170,

172. Likewise, the Green Mountain and Crow Creek grazing permits identify the

AOIs as a key source of management practices that are required of permittees and

that are incorporated into the permits. As the permits explain,

The specific management practices required of the permittee, such as
riding, salting, pasture rotations, herding, bedding, etc. are
incorporated into this permit through the approved Allotment

-40-
 Management Plan (AMP) and the Annual Operating Instructions
(AOI).

The permittee’s grazing management practices will be in compliance
with all applicable Forest Plan direction and Management Area
standard[s] and guidelines. This direction and standards/guidelines
[are] incorporated into this permit through the approved AMP and
AOI’s.

Id. Vol. 2 at 342 (Crow Creek) (emphasis added); id. at 351 (Green Mountain).

Hence, the AOIs are agency action.

The Crow Creek and Green Mountain AOIs also constitute final action.

Under the Supreme Court’s decision in Bennett v. Spear, agency action is final if

it satisfies two requirements: “First, the action must mark the consummation of

the agency’s decisionmaking process—it must not be of a merely tentative or

interlocutory nature. And second, the action must be one by which rights or

obligations have been determined, or from which legal consequences will flow.”

520 U.S. 154, 177–78 (1997) (citation and internal quotation marks omitted);

accord Pennaco Energy, Inc. v. U.S. Dep’t of Interior, 377 F.3d 1147, 1155 (10th

Cir. 2004). The Supreme Court has “interpreted the ‘finality’ element in a

pragmatic way.” FTC v. Standard Oil of Cal., 449 U.S. 232, 239 (1980) (internal

quotation marks omitted). If an agency has issued a “definitive statement of its

position, determining the rights and obligations of the parties,” the agency’s

action is final notwithstanding “[t]he possibility of further proceedings in the

agency” on related issues, so long as “judicial review at the time [would not]

-41-
disrupt the administrative process.” Bell v. New Jersey, 461 U.S. 773, 779–80

(1983); see also Sierra Club v. U.S. Army Corps of Eng’rs, 446 F.3d 808, 813

(8th Cir. 2006); cf. Sierra Club v. Peterson, 228 F.3d 559, 563, 566–67 (5th Cir.

2000) (groups’ challenge to timber-management program is not final agency

action even though group identified particular timber sales because complaints

filed in case indicated sales were simply “examples” of a general program of

timber management the groups sought to challenge that included “past, ongoing,

and future timber sales”).

As to Bennett’s first prong, the AOIs are undoubtedly the consummation of

the Forest Service’s decisionmaking process. They identify when grazing may

begin and when it will end, and which pastures may be used at particular times.

They serve as the Forest Service’s annual determinations regarding how much

grazing will be allowed each season, for they explicitly distinguish between

grazing that was “permitted” under the term grazing permit and grazing that is

actually “authorized” for a particular grazing season. See, e.g., Aplts. App. Vol.

1 at 173. The distinction between “permitted” grazing and “authorized” grazing

is significant, because, as the facts before us illustrate, the differences between

the two amounts may be substantial. And no further agency action is required to

make the AOI binding on permittees. See Bennett, 520 U.S. at 178.

We recognize that an AOI may be described as a “management tool” for the

Forest Service, and events during the grazing season (such as a fire) can require

-42-
further modifications to what grazing is permitted. But AOIs are the last word

before grazing begins and undoubtedly have clear and definite consequences for

permittees, who need to make their plans based on what the AOIs authorize. In

other words, AOIs “ha[ve] a direct and immediate effect on the day-to-day

business” of permittees, and “immediate compliance with their terms [is]

expected.” Standard Oil Co. of Cal., 449 U.S. at 239–40 (ellipsis and internal

quotation marks omitted). Moreover, the issuance of the AOIs presents a “legal

issue . . . fit for judicial resolution.” Id. at 240 (internal quotation marks

omitted). We note that the Forest Service does not contend that the reason why

the AOIs lack finality is that there may be revisions during the grazing season.

Turning to the second prong of the Bennett test, we note that the Crow

Creek and Green Mountain AOIs are actions “by which rights or obligations have

been determined, or from which legal consequences will flow.” 520 U.S. at 178

(internal quotation marks omitted). The Forest Service Handbook for the Rocky

Mountain Region explains that AOIs should “identify the obligations of the

permittee and the Forest Service . . . [and] clearly articulate annual grazing

management requirements and standards.” U.S. Forest Serv., Forest Service

Handbook Rocky Mountain Region, § 2209.13.96.3. Among other things, an AOI

sets forth “[t]he maximum permissible grazing use authorized on the allotment for

the current grazing season.” Id. If a permittee fails to comply with the

maintenance standards and management practices outlined in the AOIs, its permit

-43-
may be cancelled or suspended. See, e.g., Aplts. App. Vol. 2 at 337 (Crow Creek

grazing permit) (“[T]his permit may be suspended or cancelled . . . for failure to

comply with any of the terms and conditions in Parts 1, 2, and 3 hereof . . . .”); id.

at 341 (part 3 of grazing permit) (“Maintenance standards are identified annually

in the [AOI].”); id. at 342 (part 3 of grazing permit) (“The specific management

practices required of the permittee . . . are incorporated into this permit through

the . . . [AOI].”). The AOIs accordingly satisfy Bennett’s second prong. See also

City of Dania Beach, Fla. v. Fed. Aviation Admin., 485 F.3d 1181, 1188–89 (D.C.

Cir. 2007) (FAA letter is a “final” order because, in part, it “provides new

marching orders about how air traffic will be managed at [a particular airport]”).

Oregon Natural Desert Ass’n v. United States Forest Service (ONDA), 465

F.3d 977, 990 (9th Cir. 2006), held that AOIs issued to permittees of livestock

grazing on national forest land were final agency action. The Forest Service

contends however, that ONDA is distinguishable because its determination that

AOIs were the consummation of its decisionmaking process turned on the Forest

Service’s failure in that case to issue allotment management plans for five of the

six grazing permits at issue. See id. at 984. We do not share the Forest Service’s

interpretation. ONDA focuses largely on the fact that an AOI is the only

document that takes into account information, including drought conditions and

water quality, not available when an allotment management plan or grazing

permit is issued. See id. at 980–81, 984–85.

-44-
 The Forest Service also urges us to adopt the views of Judge Fernandez’s

dissent in ONDA, which contended that “AOIs are merely a way of conducting the

grazing program that was already authorized and decided upon when the permits

were issued.” Id. at 991 (Fernandez, J., dissenting). In Judge Fernandez’s view,

it is only the grazing permits and not the AOIs that constitute final agency action.

See id. at 990. We disagree. As his dissent acknowledged, AOIs “provide[] for

periodic changes and adjustments, as needed, for resource protection.” Id. But

such changes may be far more than insignificant “adjustments.” Here, for

instance, the 2003 Crow Creek AOIs authorized only 1559 animal months of

grazing and the 2004 AOIs authorized only 1368 animal months, even though the

grazing permits for that allotment allowed 2047 months. Declining to treat AOIs

as final agency action would insulate from review significant decisions by the

Forest Service that constitute much more than mere implementation of grazing

permits. We conclude that the 2003 and 2004 AOIs for Crow Creek and Green

Mountain constitute final agency action under the APA and turn to the merits of

CNE’s CWA claim.

2. Merits

CNE contends that the Forest Service’s issuance of AOIs in Crow Creek

and Green Mountain in 2003 and 2004 was arbitrary and capricious under the

APA and violated CWA § 313(a), 33 U.S.C. § 1323(a), because nonpoint-source

pollution in Pole Mountain had resulted in levels of fecal-coliform bacteria

-45-
violating Wyoming regulations. We disagree because, as we shall explain,

Wyoming law does not make a nonpoint-source polluter a guarantor of water-

quality compliance. Rather, because the Forest Service has implemented

Wyoming’s best management practices, it has “compl[ied] with . . . State . . .

requirements . . . respecting the control and abatement of water pollution in the

same manner, and to the same extent as any nongovernmental entity.” Id.

The CWA is intended “to restore and maintain the chemical, physical, and

biological integrity of the Nation’s waters.” Id. § 1251(a). It seeks to achieve

this aim primarily through the regulation of point sources, which are “any

discernible, confined and discrete conveyance[s] . . . from which pollutants are or

may be discharged.” Id. § 1362(14). Section 1311(a) of the CWA prohibits the

discharge of pollutants from point sources unless certain requirements are met,

see id. § 1311(a); see also id. § 1362(12) (defining “discharge of a pollutant” as

the addition of a pollutant to protected water from any point source, with the

exception of pollutants added to certain waters from vessels or floating crafts).

Discharge of a pollutant may, however, be authorized under a National Pollutant

Discharge Elimination System (“NPDES”) permit. Id. § 1342. The State of

Wyoming has authority to issue such permits within its borders. Id. § (a)(5); 40

Fed. Reg. 13,026 (Mar. 24, 1975).

The CWA’s treatment of point-source discharges differs from its treatment

of nonpoint-source pollution, which is the alleged form of pollution at issue in

-46-
this case. Indeed, the CWA does not even define nonpoint-source pollution.

(This court, however, has adopted the description that it is “‘nothing more than a

water pollution problem not involving a discharge from a point source,’” Am.

Wildlands v. Browner, 260 F.3d 1192, 1193 (10th Cir. 2001) (quoting Nat’l

Wildlife Fed’n v. Gorsuch, 693 F.2d 156, 166 n.28 (D.C. Cir. 1982)) (brackets

omitted).) Section 1311(a) does not regulate nonpoint-source pollution.

See § 1311(a) (prohibiting “discharge of any pollutant”). And whereas the CWA

requires a permitting system for point-source discharges—whether conducted by

federal or state agencies—it deals with nonpoint-source pollution merely by

“requir[ing] states to develop water quality standards for intrastate waters.”

Defenders of Wildlife v. U.S. Envtl. Prot. Agency, 415 F.3d 1121, 1124 (10th Cir.

2005); see also 33 U.S.C. § 1313(c)(2)(A); 40 C.F.R. § 130.3; id. § 131.6.

Section 1323(a), upon which CNE relies in this appeal, requires federal

agencies to comply with state and local water-quality requirements “in the same

manner, and to the same extent as any nongovernmental entity.” Congress

intended this section to ensure that federal agencies were required to “meet all

[water pollution] control requirements as if they were private citizens.” S. Rep.

No. 92-414 (1971), as reprinted in 1972 U.S.C.C.A.N. 3668, 3734. The provision

applies to activities resulting in either “discharge or runoff of pollutants.”

§ 1323(a). The parties do not contest that § 1323(a) applies to the form of

nonpoint-source pollution at issue here, so we need not decide the provision’s

-47-
outer perimeters. But see Robin Kundis Craig, Idaho Sporting Congress v.

Thomas and Sovereign Immunity: Federal Facility Nonpoint Sources, the APA,

and the Meaning of ‘In the Same Manner and to the Same Extent as Any

Nongovernmental Entity’, 30 Envtl. L. 527, 553 (2000) (“[N]on-runoff sources of

nonpoint source pollution, such as landslides, are probably not within [33 U.S.C.

§ 1323’s] waiver of sovereign immunity.”).

Wyoming water-quality regulations set limits on fecal-coliform

concentrations. See 020-080-001 Wyo. Code R. § 27. They also provide that “no

person shall cause, threaten or allow violation of a surface water quality standard

contained herein.” Id. § 1. Relying on these provisions, CNE contends that the

Forest Service has violated 33 U.S.C. § 1323(a) because fecal-coliform readings

in Pole Mountain have exceeded permissible levels. The Forest Service does not

dispute the applicability of the above state regulations. Nor does it dispute that

there have been fecal-coliform readings that have exceeded the state limit.

The central issue with respect to this claim is what constitutes compliance

with Wyoming’s water-quality requirements. The Forest Service contends that its

current implementation of best management practices (BMPs) to address the

elevated fecal-coliform readings in Pole Mountain means that it has complied

with state water-quality requirements “to the same extent as any nongovernmental

entity.” 33 U.S.C. § 1323(a). We agree. Wyoming water-quality regulations

-48-
explicitly distinguish between those nonpoint-source polluters who have

implemented BMPs and those who have not:

The numerical and narrative standards contained within these
regulations [which include the fecal-coliform standard] shall be used
to establish effluent limitations for those discharges requiring control
via permits to discharge in the case of point sources and best
management practices in the case of nonpoint sources. If no permit
or best management practice has been issued or implemented for a
pollution source the state may, in addition to other appropriate legal
action, take direct action to enforce these standards.

020-080-001 Wyo. Code R. § 5. This provision contemplates that Wyoming

“control” nonpoint-source violations of water-quality standards with the

implementation of BMPs; only if BMPs have not been implemented is nonpoint-

source pollution not under “control” and subject to state enforcement action. The

March 2000 Wyoming Nonpoint Source Management Plan Update explains how

this is done:

The Wyoming [Nonpoint Source] Program has been developed as a
voluntary program, providing guidelines for addressing nonpoint
sources of pollution by adoption of the plan and BMPs included
therein. Upon identification of water quality standards violations
occurring as a result of nonpoint sources, the [Water Quality
Division] will work with state, local, and federal management
agencies, along with private landowners and operators, to select
appropriate BMPs and to develop a plan and schedule for
implementation.

Aplees. Jt. Supp. App. Vol. 2 at 326–27.

The Wyoming water-quality rules acknowledge that BMPs, even when

implemented, may not necessarily stop nonpoint-source pollution from exceeding

-49-
water-quality standards. They define BMPs as “a practice or combination of

practices that . . . are determined to be the most technologically and economically

feasible means of managing, preventing or reducing nonpoint source pollution.”

020-080-001 Wyo. Code R. § 2(b)(v) (emphasis added). Neither the definition of

BMPs nor section 5’s enforcement standard requires that the implementation of

BMPs for nonpoint-source pollution lead to water-quality readings that meet all

applicable standards.

It is undisputed that the Forest Service has in good faith implemented and

continues to implement BMPs in Pole Mountain. Although CNE contends that

these BMPs have failed because water-quality violations have allegedly

continued, that is not the standard dictated by state regulations and the CWA.

The March 2000 Wyoming Nonpoint Source Management Update even says that

when BMPs are ineffective, the state agency will work with the polluters and

others “to identify needed BMP modifications.” Aplees. Jt. Supp. App. Vol. 2. at

327. This hardly suggests that water-quality exceedances after BMPs have been

implemented indicate that the BMP process has failed. Moreover, at the end of

the 2004 grazing season, only one location in Pole Mountain exceeded standards,

and at that location the fecal-coliform level was one-fourth the previous year’s

level. In any event, so long as BMPs have been implemented, the state agency

has no authority to take enforcement action, and the Forest Service cannot be said

-50-
to have failed to comply with state requirements “in the same manner, and to the

same extent as any nongovernmental entity.” 33 U.S.C. § 1323(a).

The above analysis also leads us to conclude that the Forest Service’s

issuance of AOIs for Crow Creek and Green Mountain was not arbitrary or

capricious. Notably, the grazing reductions in the 2003 AOIs for Crow Creek and

the 2004 AOIs for Crow Creek and Green Mountain yielded results; at the end of

2004 only one location exceeded fecal-coliform standards, and even at that

location the level was greatly reduced. The Forest Service’s ongoing

implementation of BMPs and its entry into a Memorandum of Understanding with

local conservation districts reflect a reasoned approach to elevated fecal-coliform

levels. We cannot say that there was “a clear error of judgment” in issuing the

AOIs. Olenhouse v. Commodity Credit Corp, 42 F.3d 1560, 1574 (10th Cir.

1994) (addressing arbitrary-or-capricious standard).

The Forest Service’s issuance of AOIs in Crow Creek and Green Mountain

while BMPs are being implemented was not contrary to the CWA nor arbitrary

and capricious under the APA.

III. CONCLUSION

We AFFIRM the district court’s dismissal of CNE’s petition for review.

-51-
06-1130, Center for Native Ecosystems v. Cables

BRISCOE, Circuit Judge, concurring:

I join fully in Parts I, II.A, II.B.1, II.C, and III of the majority’s opinion. I

also agree with the result reached in Part II.B.2 of the majority’s opinion, but

write separately because my reasoning with regard to the issue discussed therein

differs substantially from the majority’s.

As the majority notes in Part II.B.2 of its opinion, CNE claims that the

Forest Service violated the Endangered Species Act (ESA) by failing, after the

relevant forage-utilization standards were allegedly violated, to reinitiate

consultation with the Fish and Wildlife Service (FWS). CNE bases its claim

exclusively on the language of 50 C.F.R. § 402.16. That regulation, entitled

“Reinitiation of formal consultation,” reads as follows:

Reinitiation of formal consultation is required and shall be requested
by the Federal agency or by the Service, where discretionary Federal
involvement or control over the action has been retained or is
authorized by law and:

(a) If the amount or extent of taking specified in the incidental take
statement is exceeded;

(b) If new information reveals effects of the action that may affect
listed species or critical habitat in a manner or to an extent not
previously considered;

(c) If the identified action is subsequently modified in a manner that
causes an effect to the listed species or critical habitat that was not
considered in the biological opinion; or

(d) If a new species is listed or critical habitat designated that may be
affected by the identified action.
 Both the title and body of the regulation make abundantly clear that it

applies only in circumstances where formal consultation has already occurred. In

particular, the regulation’s use of the word “reinitiation” clearly implies that

“initiation” of formal consultation has previously occurred. Likewise, the

references in subsections (a) and (c) of the regulation to “incidental take

statements” and “biological opinions,” both of which are products of the formal

consultation process, clearly imply that formal consultation has previously

occurred.

Given this interpretation of the regulation, CNE’s “reconsultation” claim

necessarily must fail. When the Forest Service issued its Biological Assessment

in 1998, it concluded that, with certain grazing management steps in place, the

revised Allotment Management Plan (AMP) “‘m[ight] effect’ but [wa]s ‘not likely

to adversely affect’ the [Preble’s mouse] or its habitat.” Aplees. Jt. Supp. App.

Vol. 4 at 842. In light of this conclusion, the ESA merely required the Forest

Service to “informally consult” with the FWS, which it did (and the FWS agreed

with the Forest Service’s conclusion). In other words, the Forest Service was not

required by the ESA to, and in fact did not, “formally consult” with the FWS.

Thus, since formal consultation was never initiated regarding the revised AMP

(and no biological opinion was ever issued by the FWS), § 402.16 is inapplicable

here.

-2-