Western Watersheds Project et al v. United States Bureau of Land Management

Filing 55

ORDER granting in part and denying in part 36 Motion for Summary Judgment; denying 49 Cross-Motion for Summary Judgment: Plaintiffs' Motion for Summary Judgment 36 is granted to the extent that the case will be remanded to Bureau of Land Management. It is otherwise denied without prejudice to renewal. Defendant's Cross-Motion for Summary Judgment 49 is denied without prejudice to renewal. This case is remanded to the Bureau of Land Management for further proceedings consistent with this opinion. Bureau of Land Management shall file a supplemental report by 4/24/2015. Status Hearing is set for 5/18/2015 at 11:00 a.m. before Judge Rosenblatt. The parties shall file a joint status report at least fourteen days before the Status Hearing. See order for additional details. Signed by Senior Judge Paul G Rosenblatt on 2/26/2015.(LMR)

1

WO

2
3
4
5
6
7

IN THE UNITED STATES DISTRICT COURT

8

FOR THE DISTRICT OF ARIZONA

9
10
11

Western Watersheds Project, et al.,

12

Plaintiffs,

13

ORDER

v.

14

No. CV-13-01028-PHX-PGR

United States Bureau of Land Management,

15

Defendant.

16
17
18

This case involves challenges under the National Environmental Policy Act

19

(“NEPA”) to the Bureau of Land Management’s (“BLM”) decision regarding

20

management of livestock grazing on the Sonoran Desert National Monument.

21
22

Specifically, Plaintiffs Western Watershed Project and Sierra Club (collectively,

23

“WWP”) contend that the process by which BLM made its decision to allow grazing on

24

the Sonoran Desert National Monument lands north of Interstate Highway 8 (“I-8”) was

25

not adequately explained nor adequately supported by the administrative record and thus

26
27
28

violates NEPA. The parties have filed cross-motions for summary judgment. (Doc. 36
(WWP); Doc. 49 (BLM).) The Court will grant in part and deny in part summary

1

judgment in favor of Plaintiffs, and will deny BLM’s cross motion for summary

2

judgment.1

3
Background

4
5
6

A.

Establishment of the Sonoran Desert National Monument
The Sonoran Desert National Monument (the “Monument”), which was

7
8
9
10
11

established by Presidential Proclamation in 2001, covers more than 486,000 acres in
southwest Arizona. The Monument was set aside to protect the Sonoran desert landscape
and diverse plant communities, animals, and historical sites found there.

The

Proclamation characterizes the Monument as “the most biologically diverse of the North

12
13

American deserts,” with a “spectacular diversity of plant and animal species.” (AR

14

3886.) “The most striking aspect of the plant communities within the Monument are the

15

abundant saguaro cactus forests,” which are a “national treasure, rivaling those within the

16
17
18
19

Saguaro National Park.” (Id.)
The Monument was created for the purpose “of protecting the objects identified”
in the Proclamation, and thus the activities that can occur on the Monument are restricted.

20
21

(AR 3887-88.) Use of motorized and mechanized vehicles off roads is prohibited within

22

the Monument, and the Monument lands are withdrawn from any form of entry, sale,

23

lease, or other disposition, including for mining and mineral development. (Id.) Grazing

24

within the Monument is also restricted. Specifically, although the “[l]aws, regulations,

25
26
27
28

1

The Court finds that oral argument would not assist in resolving this matter and
accordingly finds the pending motions suitable for decision without oral argument. See
LRCiv 7.2(f); Fed.R.Civ.P. 78(b); Partridge v. Reich, 141 F.3d 920, 926 (9th Cir. 1998).

-2-

1
2

and policies followed by the Bureau of Land Management in issuing and administering
grazing permits and leases” continue to apply to Monument lands, the Proclamation

3
4
5
6

provides “that grazing permits on Federal lands within the Monument south of Interstate
Highway 8 shall not be renewed at the end of their current term.” (AR 3888.) Further,
on Monument lands north of I-8, grazing may continue “only to the extent that the

7
8
9
10
11

Bureau of Land Management determines that grazing is compatible with the paramount
purpose of protecting the objects identified in this proclamation.” (Id.)
B.

The Grazing Decision
To address the actions “necessary to protect the objects” of the Monument, BLM

12
13

was required to prepare a management plan. (AR 3888.) In 2002, BLM began that

14

process by gathering data for the purpose of conducting a Land Health Evaluation

15

(“LHE”).

16
17
18
19

1.

PBI Data

BLM contracted with The Nature Conservancy (“TNC”), as a primary partner in
an interdisciplinary team, to make rangeland health assessments on the Monument. TNC,

20
21

in turn, contracted with the Pacific Biodiversity Institute (“PBI”), through an assistance

22

agreement with BLM, to collect data within the Monument. TNC and PBI conducted

23

fieldwork on the Monument lands during the period 2002 to 2006, collecting data at 320

24

different plots.

25
26

PBI’s study indicated that the lower elevation communities on the Monument

27

were most impacted by livestock grazing and showed the greatest disturbance in the form

28

of low vegetation cover, low native species diversity, and high levels of non-native
-3-

1
2

species. Of those communities, the creosote-bursage desert scrub community, which is
one of the primary communities in the Monument, had the greatest level of livestock

3
4
5
6

grazing and was one of the most disturbed communities. Further, the areas around
livestock congregation areas within those communities, such as water sources, other
range developments, and cattle trails, had the most severe degradation with highly altered

7
8
9
10
11

vegetation composition and structure and altered surface soils.
2.

BLM Data

BLM had previously collected data in 1981, and collected additional data at “key
areas” of the Monument in 2003-2004, 2007, and 2009. A “key area” is a long-term

12
13

monitoring plot located in a single ecological site and plant community which is

14

representative of the community in which it is located and represents where livestock

15

grazing pressure is occurring across the management area. Key areas are a relatively

16
17
18
19

small portion of each grazing allotment “selected because of its location, proximity to
water, livestock and wildlife habitat values, and value as a long-term monitoring point.”
BLM’s key areas are typically located approximately one mile from a water

20
21

source. WWP contends that BLM locates them at this distance from a water source,

22

rather than closer to the source, to prevent collection of data in areas with the heaviest

23

impacts. (Doc. 44-1 ¶ 29.) BLM contends that it locates the key areas approximately one

24

mile from a water source “to prevent arbitrarily skewing the data toward heavy impacts

25
26

(less than 0.5 miles from water) or toward fewer impacts (greater than 1.5 miles from

27

water).” (AR 74612; Doc. 51 ¶ 29.)

28

-4-

1
2

3.

Land Health Evaluation

BLM used the data gathered by itself and PBI to prepare the LHE, an essential

3
4
5
6

step in the process of determining whether livestock grazing is compatible with
protecting the objects of the Monument. The LHE gauged whether “Standard One” and
“Standard Three” of the Arizona Standards for Rangeland Health were being achieved on

7
8

the grazing allotments located on Monument lands north of I-8 and, if not, whether

9

livestock grazing was the causal factor.2 Only BLM’s application of Standard Three –

10

which examines desired resource conditions based on production and diversity of native

11

plant communities by assessing plant composition, structure, and cover – is at issue in

12
13

this case.

14
15

a.

Desired Plant Community Objectives

To assess Standard Three, BLM first established desired plant community

16
17
18
19

objectives for each of the “ecological sites” on the Monument. The Monument contains
seven categories of “ecological sites”: sandy wash, loamy swale, limy fan, limy upland
deep, limy upland, granitic hills, and sandy loam deep. The desired plant objectives for

20
21
22
23
24

each ecological site relate to vegetation canopy cover, vegetation composition, and, for
some ecological sites, recruitment of saguaros.
To establish the desired plant community objectives for each ecological site, BLM
considered data from corresponding ecological sites on the Barry M. Goldwater Range

25
26

and Area A (“BGR/A”), and information from Natural Resources Conservation Service

27
28

2

Arizona Standard Two, which pertains to riparian-wetland sites, was deemed to
not be applicable because none of the six allotments contain riparian areas or wetlands.
-5-

1
2

ecological site descriptions and reference sheets (collectively “ESD”).

The BGR/A

historically was grazed by livestock until the early 1940s. Since that time, BGR/A has

3
4
5
6

been closed to grazing, although it has continued to have some use by trespass livestock,
especially along the edges of the area.
b.

7

Standard Three Assessment – Whether Desired Plant Community
Objectives Were Being Met

8

After setting the desired plant community objectives, BLM compared data

9
10

collected on the six grazing allotments north of I-8 with the desired plant community

11

objectives to assess whether each of the ecological sites on each of the grazing allotments

12

was meeting the plant community objectives. BLM relied on both its own data from 36

13

plots and PBI data from 48 plots. BLM did not use any PBI or other data from plots that

14
15

were close to livestock congregation areas and thus excluded data from the heaviest

16

impact areas.

17

When comparing the key area data to the desired plant community

objectives, BLM considered a plot to be meeting the objective if the data value was

18
19

within 80% of the objective, i.e., was within the “80% threshold.”

20

The final determination of whether an ecological site within an allotment was

21

achieving desired plant community objectives was based on a preponderance of the

22
23

evidence standard.

Under this standard, if more than half of the plots within an

24

ecological site were meeting all objectives, the whole ecological site was considered to be

25

meeting objectives. This approach was used because there were not enough plots to

26

complete a valid statistical analysis for each ecological site.

27
28

In the final LHE, BLM determined that 127,550 acres, or 50.5 percent, of

-6-

1
2

Monument lands north of I-8 were not meeting the desired plant community objectives
(i.e., were not achieving Standard Three).

3
4
5
6

c.

Causality Determination

In 2009, BLM collected information to determine livestock use levels during the
2008-2009 grazing season on the allotments north of I-8. BLM used two methods to

7
8
9
10
11

assess livestock use. One method was to conduct utilization transects on the Bighorn and
Conley allotments in spring and summer 2009 to estimate percent use of certain perennial
shrub species.
The other method used by BLM was use pattern mapping on the Monument in

12
13

March and April 2009, to determine the proportion of vegetation production that had

14

been consumed or destroyed by the animals. To map the use patterns, BLM drove along

15

roads on the northern portion of the Monument and stopped every half mile to a mile to

16
17
18
19

assess livestock use of key perennial forage species. The level of livestock use was
classified on the initial use pattern map as “negligible,” “slight,” “light,” “moderate,”
“heavy,” “severe,” or “very severe.” The initial use pattern map also designated areas

20
21

that were “unsuitable” for grazing because they were too steep, and areas that were

22

“unsurveyed.” For areas with use classified as “heavy” or “severe,” BLM also conducted

23

utilization transects to verify the accuracy of the classifications.

24

The final use pattern map produced by BLM changed the designation of certain

25
26

areas from the designation that was used on the initial use pattern map. For example,

27

areas that were initially designated as “unsuitable” were changed to a designation of

28

“unsurveyed or inaccessible” on the final map, and areas that were initially designated as
-7-

1
2

“unsurveyed” were changed to a designation of “negligible” or “slight” use on the final
map. (AR 83693, 83691, 84023; Doc. 44-1 ¶ 63; Doc. 51 ¶ 63.)

3
4
5
6

BLM assumed that in areas that had greater than 40% shrub utilization (classified
as moderate, heavy, or severe use), livestock grazing was the causal factor in not
achieving the rangeland health standards. However, if areas had less than 40% shrub

7
8

utilization (classified by BLM as negligible to slight livestock use), BLM assumed that

9

livestock grazing was not the causal factor for non-achievement of the health standards.

10

In areas where BLM determined that current grazing was not the causal factor for non-

11

achievement, BLM attributed the failure to other factors such as historic livestock

12
13

grazing, historic livestock patterns, fire, drought, off-road vehicle use, or general

14

recreation.

15

BLM determined that of the Monument lands north of I-8 that were not achieving

16
17
18
19

the desired plant community objectives, livestock grazing was the causal factor on 8,498
acres. The remaining acreage of the Monument lands north of I-8 was deemed by BLM
to be compatible with livestock grazing.

20
21

Standard of Review

22

WWP brings this challenge under NEPA, which requires federal agencies to

23

consider the consequences of their actions on the environment. NEPA’s mandate is

24

“essentially procedural. . . . . It is to ensure a fully informed and well considered

25
26

decision.” Vermont Yankee Nuclear Power Corp. v. NRDC, 435 U.S. 519, 558 (1978).

27

“NEPA itself does not mandate particular results, but simply prescribes the necessary

28

process.” Robertson v. Methow Valley Citizens Council, 490 U.S. 332, 350 (1989). The
-8-

1
2

goals of NEPA are to ensure the agency will have detailed information on significant
environmental impacts when it makes its decisions and guarantee that this information

3
4
5
6

will be available to a larger audience. Id. at 349.
Because NEPA does not provide its own standard of review, the Court’s review is
governed by the Administrative Procedures Act (“APA”). Under the APA, an agency’s

7
8
9
10
11

decision can be set aside if it is “arbitrary, capricious, an abuse of discretion, or otherwise
not in accordance with law.” 5 U.S.C. § 7006(A); see Marsh v. Oregon Natural Res.
Council, 490 U.S. 360, 375 (1989). Review under the arbitrary and capricious standard is
narrow, and this Court is not to substitute its own judgment for that of the agency. See

12
13
14
15

Lands Council v. McNair, 537 F.3d 981, 987 (9th Cir. 2008) (en banc).
While agency decisions are granted deference, such decisions are not spared a
“thorough, probing, in-depth review.” Citizens to Preserve Overton Park v. Volpe, 401

16
17
18
19

U.S. 402, 415 (1971); see Marsh, 490 U.S. at 378 (A court’s review of administrative
action “must be searching and careful,” though “the ultimate standard of review is a
narrow one.”) (internal quotations omitted). To withstand such review, an agency must

20
21

have considered the relevant information and provided a satisfactory explanation for its

22

actions, drawing a “rational connection between the facts found and the choice made.”

23

Burlington Truck Lines v. United States, 371 U.S. 156, 168 (1962). “An agency’s

24

decision is arbitrary and capricious if it fails to consider important aspects of the issue

25
26

before it, if it supports its decisions with explanations contrary to the evidence, or if its

27

decision is either inherently implausible or contrary to governing law.” In Defense of

28

Animals, Dreamcatcher Wild Horse and Burro Sanctuary v. United States Dep’t of
-9-

1

Interior, 751 F.3d 1054, 1061 (9th Cir. 2014).

2

Discussion

3
WWP challenges the process used by BLM to reach its decision regarding grazing

4
5
6

on the Monument lands north of I-8. Specifically, WWP challenges BLM’s conduct of
the LHE, which provided the foundation for BLM’s grazing analysis, including its

7
8

grazing compatibility determination. WWP contends BLM failed to provide underlying

9

data or explain how the data supported its conclusions; excluded relevant information and

10

provided inconsistent or inadequate explanations for the exclusion; and failed to provide

11

a rational connection between the facts in the record and BLM’s conclusions. WWP

12
13

challenges three steps in the LHE process: (1) the identification of the desired plant

14

community objectives, (2) the determination of whether the identified objectives are

15

being met, and (3) the determination of whether livestock grazing caused the failure to

16
17
18
19

meet objectives (the “causality determination”).
A.

Identification of Desired Plant Community Objectives
WWP contends that BLM failed to adequately explain how it derived the desired

20
21

plant community objectives. BLM contends that it did provide an adequate explanation

22

for how it set the desired plant community objectives. The pages of the administrative

23

record cited by BLM explain that it set objectives for each ecological site based on the

24

average value of data from the BGR/A, or at the low end of the ESD (see AR 74601,

25
26

74608, 73126).

27

explanation in the record for adjustments it made to the BGR/A and the ESD data in

28

However, as discussed below, BLM did not provide an adequate

setting the objectives.
- 10 -

1
2

1.

Ecological Sites based on BGR/A Data

BLM had BGR/A data for four of the ecological sites: the limy fan, sandy wash,

3
4
5
6

limy upland deep, and granitic hills ecological sites. BLM stated that it set the final
objectives for each of these ecological sites based on the average BGR/A data value for
each attribute. (AR 74608.) However, a comparison of the BGR/A data and BLM’s final

7
8

objectives for the various ecological sites demonstrates that, although the granitic hills

9

final objectives for canopy cover and composition are consistent with the average BGR/A

10

data, the final objectives for the limy fan, sandy wash, and limy upland deep ecological

11

sites are not consistent with the average BGR/A data. BLM provides an explanation in

12
13

its brief for the variance between the objectives and the BGR/A data, but this explanation

14

is not contained in the administrative record3 and, further, does not adequately explain or

15

account for the wide variance in the adjustments made by BLM.

16
17
18
19

2.

Ecological Sites based on the ESD

For the three ecological sites without BGR/A data – loamy swale, limy upland,
and sandy loam deep – BLM based its objectives on the low end of the ESD. In its brief,

20
21

BLM explains that it further adjusted the ESD objectives for the lower precipitation north

22

of I-8.

23

administrative record and, further, does not adequately explain or account for the wide

24

This “lower precipitation” explanation is not, however, contained in the

variance in the adjustments made by BLM.

25
26

In addition, the LHE contains objectives for “palatable browse” for the limy

27
28

3

The peer review comment cited by BLM in its brief is contained in the record
(see AR 55411), but does not provide the type of explanation required under NEPA.
- 11 -

1
2

upland site and sandy loam site, and an objective for “perennial grass composition” for
the loamy swale site. However, the ESD does not include any objectives for “palatable

3
4
5
6

browse” or “perennial grass composition.” In its brief, BLM provides explanations for
how it derived both of these objectives, but, again, these explanations are not contained in
the administrative record.

7
3.

8

Saguaro Recruitment

9

BLM set final saguaro recruitment objectives for only two ecological sites – the

10

limy upland site and the granitic hills site – and based those objectives on the saguaro

11

recruitment rate in the BGR/A. BLM explains in its brief that in setting the objectives, it

12
13

adjusted downward from the BGR/A data due to lower precipitation or other variance in

14

the ecological sites north of I-8. This explanation is not, however, contained in the

15

administrative record and, further, does not explain the wide variance in the adjustments

16
17
18
19

made.
4.

Conclusion

In sum, BLM did not provide an adequate explanation in the record to support its

20
21

setting of, and/or adjustments to, the objectives for the limy fan, sandy wash, limy upland

22

deep, and granitic hills ecological sites.

23

objectives was therefore arbitrary and capricious. See Humane Soc'y of U.S. v. Locke,

24

BLM’s setting of these plant community

626 F.3d 1040, 1049 (9th Cir. 2010) (“We cannot gloss over the absence of a cogent

25
26

explanation by the agency by relying on the post hoc rationalizations offered by

27

defendants in their appellate briefs.”); Northwest Coalition for Alternatives to Pesticides

28

(NCAP) v. U.S. E.P.A., 544 F.3d 1043, 1052 n.7 (9th Cir. 2008) (“where the agency’s
- 12 -

1
2

reasoning is irrational, unclear, or not supported by the data it purports to interpret, we
must disapprove the agency’s action”).

3
4
5
6

B.

Determination of Whether Desired Plant Community Objectives Being Met
After BLM set the desired plant community objectives, the next step for BLM was

an assessment of the conditions on the allotments north of I-8 for a determination of

7
8
9
10
11

whether the allotments were meeting those objectives. To make this assessment, BLM
gathered data from the allotments north of I-8, and compared that data to the objectives.
WWP contends that BLM’s process of determining whether the allotments were
meeting the objectives was arbitrary and capricious because BLM omitted relevant data

12
13

and either failed to justify the omission or provided contradictory and inconsistent

14

explanations for the omission; and failed to adequately explain or justify the use of an

15

“80% threshold” and a “preponderance of the evidence” standard.

16
17
18
19

1.

Exclusion of BLM data

In the early versions of the LHE, BLM included all of the data it had collected to
that point in time, including data from 1981, 2004, 2007, and 2009, to assess whether

20
21

ecological sites on the allotments north of I-8 met the objectives. However, in the final

22

LHE, BLM removed all prior data and used only the most recent 2009 data from the

23

allotments, except for the Beloat allotment, for which BLM used 20044 data, even though

24
25
26
27
28

4

BLM admitted in its response brief and in its statement of controverting facts that
it used 2004 data for the Beloat allotment rather than 2009 data (see Doc. 51 ¶ 44).
However, BLM contends for the first time in its reply brief that it misidentified certain
data for the Beloat allotment in the LHE as being collected in 2004. BLM has not,
however, moved to amend its statement of controverting facts to correct this purported
- 13 -

1
2

it had 2007 and 2009 data for the allotment.
BLM explained that it used the best data available for its analysis, but did not

3
4
5
6

explain why it used 2009 data for five of the allotments, but then used 2004 data for the
Beloat allotment, even though it had available data from 2007 and 2009 for that
allotment, nor did BLM explain why the 2004 data was considered to be the “best

7
8

available data” over the more recent data. Because BLM has not provided an adequate

9

explanation for its use of the 2004 data, rather than more recent data, as an internal BLM

10

reviewer noted, this gives the appearance that BLM was “cherry-picking” data (AR

11

83853).

12
13

2.

14

BLM admits that it used some of the PBI data for its analysis, and excluded other

15

Exclusion of PBI data

PBI data, but contends that where it declined to use PBI data, it provided a reasonable

16
17
18
19

explanation for doing so.
a.

Data from heavier impact areas close to livestock waters

BLM excluded from its analysis PBI data from areas less than a mile from

20
21

livestock water and other congregation areas, which were the most heavily impacted

22

areas. BLM explained that it used existing methods to measure the impacts of grazing in

23

“key areas.” (AR 3857, 74357, 74612, 55414, 78494). In particular, “key areas” were

24

“selected in locations that represent where livestock grazing pressure is occurring” and

25
26

are located “approximately 1 mile (+/- 0.5 miles) from a water source to prevent

27
28

error. Moreover, BLM does not deny that it used 2004 data for some of the Beloat
allotment, even though it had 2007 and 2009 data available.
- 14 -

1
2

arbitrarily skewing the data toward heavier impacts (less than 0.5 miles from water) or
toward fewer impacts (greater than 1.5 miles from water).” (AR 74612.) BLM also

3
4
5
6

explained that it did not completely ignore the high impact on livestock water areas, but
instead included consideration of those impacts on Monument objects in the FEIS and in
the use pattern mapping.

7
8

The Court finds that BLM provided an adequate explanation for its exclusion of

9

the PBI data from areas less than a mile from livestock water and other congregation

10

areas, and, further, that BLM took the requisite “hard look” at, and provided a

11

“reasonably thorough discussion of,” these most heavily impacted areas. See Okanogan

12
13

Highlands Alliance v. Williams, 236 F.3d 468, 473 (9th Cir. 2000); see also Northern

14

Plains Resource Council, 668 F.3d at 1085 (NEPA requires agency to “carefully consider

15

information about significant environmental impacts”).

16
17
18
19

b.

Exclusion of data because it failed to incorporate certain factors

BLM also excluded certain PBI data on the ground that the PBI data did not
address factors such as livestock intensity, frequency, timing, season of use, or

20
21

precipitation patterns. (AR 74108, 74348-49).

However, a review of the record

22

demonstrates that BLM’s own plot data did not include such information, and BLM did

23

not incorporate that information into its LHE analysis. (See AR 55485-96, 74666-82).

24

Although the LHE report contained data about actual livestock use and precipitation, that

25
26

data was not used to determine whether ecological sites met objectives. (AR 74616-18,

27

74622-43.) Instead, the determination of whether ecological sites met objectives was

28

simply based on comparing the plot data to the objectives for plant cover, composition,
- 15 -

1
2

and saguaro recruitment, and information on livestock intensity, frequency, timing,
season of use, or precipitation was neither incorporated into nor considered in this

3
4
5
6

determination.
Moreover, BLM relied on not only its own data from 36 BLM plots in conducting
its assessment, but also on data from 48 PBI plots. (AR 74613, 74615.) BLM did not

7
8

provide a reasonable explanation as to why it rejected some PBI data based on failure to

9

incorporate factors such as livestock intensity, frequency, timing, season of use, or

10

precipitation pattern factors, but then used other PBI data even though that data also did

11

not incorporate the same factors. Absent such explanation, BLM’s decision to exclude

12
13

the PBI data was arbitrary and capricious. See In Defense of Animals, 751 F.3d at 1061;

14

NCAP, 544 F.3d at 1052 n.7.

15

c.

Exclusion of data because it was only a single year’s data.

16
17
18
19

Another reason given by BLM for excluding most of the PBI data from its analysis
was that the PBI data was only from a single year, and a single year’s data was not
enough to support sound conclusions. (AR 62582, 74107-08, 74348-49.) Yet, BLM used

20
21

only a single year’s data in its key area analysis, either from 2004 or 2009. Thus,

22

although BLM stated that a single year’s data is not enough to support sound conclusions,

23

it chose to rely on just a single year’s data for its analysis.

24

BLM does not provide any explanation for why it was acceptable to use just a

25
26

single year’s data for its determination of whether the objectives were being met, but then

27

reject the PBI data in part because a single year’s data is not enough to support sound

28

conclusions. This lack of any reasonable explanation for the inconsistent treatment of
- 16 -

1
2

data renders BLM’s decision to exclude the PBI data arbitrary and capricious. See In
Defense of Animals, 751 F.3d at 1061; NCAP, 544 F.3d at 1052 n.7.

3
3.

4

When comparing the 2009 or 2004 plot data to the desired plant community

5
6

Use of Eighty Percent Threshold Approach

objectives, BLM considered the plot to be meeting an objective if the data value was

7
8

within 80% of the objective (the “80% threshold”). BLM explained that it used the 80%

9

threshold to account for site variability when determining whether an objective was being

10

met in an individual plot (AR 74621). Thus, if a key area attribute, such as canopy cover,

11

was within 80% of the objective for that ecological site, the objective was considered to

12
13

have been met for that plot. (Id.)

14
15

WWP challenges BLM’s use of this 80% threshold on two grounds. First, WWP
contends that BLM failed to provide an adequate explanation on the source of the 80%

16
17
18
19

value.

Second, WWP contends that BLM already accounted for variability in the

ecological sites by using the average value for canopy cover or composition from the
BGR/A data, and that the use of the 80% threshold in addition to the use of the average

20
21

BGR/A data resulted in BLM “doubly accounting for site variability.”

22

As to the source of the 80% value, BLM did not explain in the LHE why it chose

23

that value (AR 74620), but in its response to comments, BLM stated that peer reviewers

24

had suggested using a range around the absolute value rather than the absolute value itself

25
26

to judge achievement of objectives to better represent real conditions on the ground, and

27

that the 80% figure was based on best professional judgment (AR 74336). In addition, in

28

response to protests, BLM stated:
- 17 -

1
2
3
4
5
6

The BLM determined that 80 percent was an appropriate threshold based on the
professional judgment of BLM specialists. Specifically, the BLM judged that most
of the variation in canopy cover and vegetation composition on Sonoran Desert
ecological sites that had not been grazed by livestock for nearly 70 years, would
lie within 80 percent or greater of the average value.
(AR 78491.)
As to the contention that BLM was doubly accounting for site variability, as BLM

7
8
9
10
11

points out, the draft LHE received by peer reviewers already incorporated the average
value for canopy cover or composition from the BGR/A data.

(See AR 55117-20

(10/19/09 draft of LHE); AR 55401 (reviewer comments, referencing 10/19/09 letter to
peer reviewers).) Three of the peer reviewers expressed concern that the objectives did

12
13

not adequately account for site variability. (See AR 55401, 55410, 55416, 83941.) To

14

address this concern, BLM added the 80% threshold to account for site variability. (AR

15

74621 n.**.)

16
17
18
19

These explanations by BLM regarding the use of the 80% threshold are both
reasonable and supported by the administrative record.
4.

Use of Preponderance of the Evidence Approach

20
21

BLM used a preponderance of the evidence approach to determine whether

22

objectives were achieved within an ecological site on an allotment. Under this approach,

23

if more than half of the plots within a given ecological site on an allotment were

24

achieving all of the plant community objectives, the ecological site was determined to be

25
26
27
28

achieving land health standards. (AR 74621.)
WWP challenges BLM’s use of the preponderance of the evidence approach,
contending that BLM did not explain how it came up with the approach. Specifically,
- 18 -

1
2

WWP contends that BLM did not explain why it chose a method where as long as 51% of
the plots on an ecological site on an allotment were meeting the objectives, the entire

3
4

ecological site was deemed to be meeting land health standards.
However, BLM explained that it used the preponderance of the evidence approach

5
6

because there were not a statistically significant number of plots within each ecological

7
8
9

site (AR 74621). In addition, a peer reviewer explained why such an approach was
needed:

10

The [LHE] provides neither statistical analysis nor discussion of data variability.
In my opinion, it is not practical to expect landscape level management analyses to
develop statistically referenced data. Using a preponderance of evidence
approach, combined with the professional judgment of experienced BLM
personnel is the best that can be reasonably be accomplished.

11
12
13
14

(AR 55409-10.)

15

BLM’s use of the preponderance approach, in light of its explanation and the peer

16
17

reviewer’s comment, is not “so implausible that it could not be ascribed to a difference in

18

view or the product of agency expertise.” WildEarth Guardians v. U.S. E.P.A., 759 F.3d

19

1064, 1070 (9th Cir. 2014).

20
21

C.

Determination of Whether Grazing Caused Failure to Meet Objectives – the
Causality Determination

22
23

Utilizing the 80% threshold and the preponderance of the evidence approach,

24

BLM concluded that 127,550 acres, or 50.5% of the Monument lands north of I-8, were

25

not achieving the land health standards and thus not achieving Standard Three. The next

26

step for BLM was to determine whether the failure to achieve the land health standards

27
28

was caused by livestock use. To make this determination, BLM used two methodologies

- 19 -

1
2

– utilization data and pattern mapping. Using these methods, BLM determined that of the
128,500 acres that were failing standards in the Monument lands north of I-8, livestock

3
4
5
6

grazing was the causal factor on only 8,498 acres. (AR 74644.)
As BLM correctly contends, its choice of methodologies is entitled to deference.
See Marsh, 490 U.S. at 377. However, WWP is not challenging BLM’s choice of

7
8
9
10
11

methodologies but is instead challenging the application of those methodologies.
Specifically, WWP contends that BLM did not implement or use the methodologies in a
rational way.
1.

Utilization Data

12
13

BLM collected utilization data in 2009 on four of the six allotments north of I-8 to

14

determine livestock use levels during the 2008-2009 grazing season. Precipitation for

15

2008 was slightly above average and there was moderate ephemeral production (400 to

16
17
18
19

500 pounds per acre) that allowed for livestock use of annual plants.5 WWP contends
that BLM’s reliance on just the 2009 utilization data to determine causality was irrational.
Further, WWP contends that BLM never responded to a peer reviewer comment that

20
21

using only data collected in 2009 was insufficient for the causality determination, and

22

that using only that single year’s data cannot account for the long-term effects to

23

vegetation, or use patterns, that might occur in non-ephemeral years when livestock are

24

grazing more perennial plants.

25
26

BLM contends that its use of a single year’s data to determine whether a site was

27
5

28

Ephemeral production, which varies with the level of precipitation, can reach as
high as 2,000 pounds per acre on the highest precipitation years, but is minimal on low
precipitation years. (AR 74619; see AR 74616-18.)
- 20 -

1
2

meeting objectives is supported by the record because the data used, from 2009 (except
for the Beloat allotment, which used 2004 data) was determined by BLM to be the “best

3
4
5
6

available data,” that the forage production for these years was average or slightly above
average, and that there is no indication that actual use in 2008-2009 was significantly
different from average. BLM further contends that it did respond to the peer reviewer

7
8
9
10
11

comment. BLM has not, however, cited to information in the administrative record that
provides an adequate explanation for its reliance on only a single year’s utilization data or
an adequate response to the concerns expressed in the peer reviewer’s comments.
First, the explanation that utilization measures a single year’s use of forage does

12
13

not respond to the peer reviewer’s concern that utilization data from only one year was

14

relied on by BLM. In other words, the fact that utilization data measures a single year’s

15

utilization does not address peer reviewer’s concern that BLM did not consider utilization

16
17
18
19

data from multiple years rather than a single year.
Second, BLM did not address the peer reviewer’s concern that BLM’s use of only
a single year’s utilization data from 2009 – which had ephemeral production that could be

20
21

utilized by the livestock – did not account for the long-term effects to vegetation, or use

22

patterns, that might occur in non-ephemeral years in which livestock graze on more

23

perennial plants.

24

Finally, as discussed previously, BLM stated it could not rely on PBI study data in

25
26

part because “one year of PBI data, in itself is not enough to support sound conclusions.”

27

(AR 74107.) Yet, BLM relied on one year of its own data to support its conclusion of

28

whether livestock use was the cause of the non-achievement of land health standards.
- 21 -

1
2

BLM has provided no explanation as to why it was justifiable to rely on a single year’s
data to determine utilization in light of its exclusion of PBI data in part because a single

3
4
5
6

year’s data is not enough to support sound conclusions.
BLM’s reliance on a single year of utilization data without providing a reasonable
explanation or addressing the concerns raised by peer reviewers rendered its causality

7
8

determination arbitrary and capricious.

See Western Watersheds Project v.

9

Kraayenbrink, 632 F.3d 472, 492-93 (9th Cir. 2011) (failure to assess, consider, and

10

respond to serious public comments can render agency decision arbitrary and capricious);

11

see also In Defense of Animals, 751 F.3d at 1061.

12
13

2.

14

The second method used by BLM to make its causation determination was to

15

Use Pattern Mapping

conduct use pattern mapping on the Monument in March and April 2009. This method

16
17
18
19

maps the proportion of vegetation production that has been consumed or destroyed by the
animals.
On the original map produced in March 2009, unsurveyed areas were marked as

20
21

“not surveyed,” while unsuitable areas were marked as “unsuitable,” and areas that did

22

not have key forage species were marked as “no forage species.” In the final version of

23

the use pattern map, BLM made many changes to the classifications. Areas that had been

24

designated as “unsurveyed” or “no forage species” on the original map were changed to

25
26

the designation of “negligible” use or “slight” use in the final version of the map, and

27

high elevation areas that were designated as “unsuitable” on the original map had been

28

changed to “unsurveyed or inaccessible.”
- 22 -

1
2

BLM does not deny that it made these changes to the final map, but contends that
it estimated use in the unsurveyed areas in later versions of the map by using its

3
4
5
6

professional judgment, combined with quantitative utilization measurements conducted at
key area and PBI plots, and distance from water sources. In support of its contention,
BLM provides various citations to the record. (AR 78529 (listing factors taken into

7
8

account in response to protest); AR 73490 (“In the absence of quantitative data, best

9

professional judgment is used”); AR 74649 (map showing key areas and PBI plots in

10

wilderness areas and water sources); AR 74650 (use pattern map showing water sources).

11

BLM also noted that livestock use is slight to negligible in steep, rocky terrain, far from

12
13

available water, which is characteristic of wilderness areas (AR 74156). Thus, BLM

14

contends that, based on its professional judgment and quantitative utilization

15

measurements, it reasonably estimated use of livestock in wilderness areas in later drafts

16
17
18
19
20
21
22
23
24
25

and in the final utilization map. BLM also cites to language located at AR 78529, which
states:
In accordance with guidance provided in TR 1743-3, the BLM considered certain
factors such as topography, rockiness, size of the area, location of salt, and
distance from water when conducting the Use Pattern Mapping because all of
these factors affect foraging habitats. The areas identified on the use pattern field
map as unsurveyed or inaccessible to livestock reflect large areas of slight to no
use due to the factors identified above.
(AR 78529.)
The Court has reviewed the citations to the record provided by BLM, as well as

26
27

the pattern use maps, and concludes that the evidence in the record provides an adequate

28

explanation for the changes to the map, and that, accordingly, BLM’s final use pattern

- 23 -

1

map, and reliance on that map, was reasonable.

2

Conclusion

3
4
5
6
7
8
9
10
11

For all of the reasons stated above, the Court concludes that BLM has failed to
adequately explain some of its decisions that led to the LHE and compatibility
determinations, and failed to address significant concerns raised in a peer reviewer’s
comments.6 The LHE is therefore “arbitrary, capricious, an abuse of discretion, or
otherwise not in accordance with law.” 5 U.S.C. § 706(2).
However, it appears that BLM may be able to readily cure the defects in its
decision-making process if given an opportunity to do so. Accordingly, the Court finds

12
13

that this is one of those “rare circumstances” in which remand without vacatur is

14

appropriate. See Humane Society of United States v. Locke, 626 F.3d 1040, 1053 & n.7

15

(9th Cir. 2010); Idaho Farm Bureau Fed'n v. Babbitt, 58 F.3d 1392, 1405–06 (9th Cir.

16
17
18
19

1995) (remanding without vacatur appropriate “when equity demands” it); see also
Heartland Regional Medical Ctr. V. Sebelius, 566 F.3d 193, 198 (C.A.D.C. 2009)
(remand without vacatur may be appropriate “[w]hen an agency may be able readily to

20
21

cure a defect in its explanation of a decision”). The Court will therefore grant summary

22

judgment in favor of WWP to the extent it seeks remand to the BLM, but will deny

23

summary judgment, without prejudice to renewal, to the extent WWP seeks to vacate

24

BLM’s decision. The Court will deny BLM’s motion for summary judgment without

25
26

prejudice to renewal. The Court will remand to BLM to provide it with the opportunity

27
28

6

The Court has considered and rejects other arguments raised by the parties that
have not been explicitly addressed in this order.
- 24 -

1
2

to either articulate reasoned explanations and responses, or adopt different decisions with
reasoned explanations that support them.

3
4
5
6

IT IS ORDERED that Plaintiffs’ Motion for Summary Judgment (Doc. 36) is
Granted to the extent that the case will be remanded to Bureau of Land Management. It
is otherwise Denied without prejudice to renewal.

7
8
9
10
11

IT IS FURTHER ORDERED that Defendant’s Cross-Motion for Summary
Judgment (Doc. 49) is Denied without prejudice to renewal.
IT IS FURTHER ORDERED that this case is Remanded to the Bureau of Land
Management for further proceedings consistent with this opinion.

12
13

IT IS FURTHER ORDERED that Bureau of Land Management shall file a

14

supplemental report, providing the required reasoned explanations and responses, or

15

indicating that it will be adopting different decisions with reasoned explanations that

16
17
18
19

support them, by April 24, 2015.
IT IS FURTHER ORDERED that a Status Hearing is set for May 18, 2015, at
11:00 a.m., in Courtroom 601 of the Sandra Day O’Connor United States Courthouse,

20
21

401 W. Washington Street, Phoenix, Arizona 85003.

22

IT IS FURTHER ORDERED that the parties shall file a joint status report at least

23

fourteen days before the Status Hearing discussing the status of the case and how this

24

case should proceed.

25
26

Dated this 26th day of February, 2015.

27
28

- 25 -