ORDER

PAUL G. ROSENBLATT, District Judge.

This case involves challenges under the National Environmental Policy Act ("NEPA") to the Bureau of Land Management's ("BLM") decision regarding management of livestock grazing on the Sonoran Desert National Monument. Specifically, Plaintiffs Western Watershed Project and Sierra Club (collectively, "WWP") contend that the process by which BLM made its decision to allow grazing on the Sonoran Desert National Monument lands north of Interstate Highway 8 ("I-8") was not adequately explained nor adequately supported by the administrative record and thus 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 judgment in favor of Plaintiffs, and will deny BLM's cross motion for summary judgment.1

Background

A. Establishment of the Sonoran Desert National Monument

The Sonoran Desert National Monument (the "Monument"), which was 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 American deserts," with a "spectacular diversity of plant and animal species." (AR 3886.) "The most striking aspect of the plant communities within the Monument are the abundant saguaro cactus forests," which are a "national treasure, rivaling those within the 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. (AR 3887-88.) Use of motorized and mechanized vehicles off roads is prohibited within the Monument, and the Monument lands are withdrawn from any form of entry, sale, lease, or other disposition, including for mining and mineral development. (Id.) Grazing within the Monument is also restricted. Specifically, although the "[l]aws, regulations, 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 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 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 was required to prepare a management plan. (AR 3888.) In 2002, BLM began that process by gathering data for the purpose of conducting a Land Health Evaluation ("LHE").

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, in turn, contracted with the Pacific Biodiversity Institute ("PBI"), through an assistance agreement with BLM, to collect data within the Monument. TNC and PBI conducted fieldwork on the Monument lands during the period 2002 to 2006, collecting data at 320 different plots.

PBI's study indicated that the lower elevation communities on the Monument were most impacted by livestock grazing and showed the greatest disturbance in the form of low vegetation cover, low native species diversity, and high levels of non-native 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 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 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 monitoring plot located in a single ecological site and plant community which is representative of the community in which it is located and represents where livestock grazing pressure is occurring across the management area. Key areas are a relatively 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 source. WWP contends that BLM locates them at this distance from a water source, rather than closer to the source, to prevent collection of data in areas with the heaviest impacts. (Doc. 44-1 ¶ 29.) BLM contends that it locates the key areas approximately one mile from a water source "to prevent arbitrarily skewing the data toward heavy impacts (less than 0.5 miles from water) or toward fewer impacts (greater than 1.5 miles from water)." (AR 74612; Doc. 51 ¶ 29.)

3. Land Health Evaluation

BLM used the data gathered by itself and PBI to prepare the LHE, an essential 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 the grazing allotments located on Monument lands north of I-8 and, if not, whether livestock grazing was the causal factor.2 Only BLM's application of Standard Three — which examines desired resource conditions based on production and diversity of native plant communities by assessing plant composition, structure, and cover — is at issue in this case.

a. Desired Plant Community Objectives

To assess Standard Three, BLM first established desired plant community 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 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 and Area A ("BGR/A"), and information from Natural Resources Conservation Service 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 been closed to grazing, although it has continued to have some use by trespass livestock, especially along the edges of the area.

b. Standard Three Assessment — Whether Desired Plant Community Objectives Were Being Met

After setting the desired plant community objectives, BLM compared data collected on the six grazing allotments north of I-8 with the desired plant community objectives to assess whether each of the ecological sites on each of the grazing allotments was meeting the plant community objectives. BLM relied on both its own data from 36 plots and PBI data from 48 plots. BLM did not use any PBI or other data from plots that were close to livestock congregation areas and thus excluded data from the heaviest impact areas. 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 within 80% of the objective, i.e., was within the "80% threshold."

The final determination of whether an ecological site within an allotment was achieving desired plant community objectives was based on a preponderance of the evidence standard. Under this standard, if more than half of the plots within an ecological site were meeting all objectives, the whole ecological site was considered to be meeting objectives. This approach was used because there were not enough plots to complete a valid statistical analysis for each ecological site.

In the final LHE, BLM determined that 127,550 acres, or 50.5 percent, of Monument lands north of I-8 were not meeting the desired plant community objectives (i.e., were not achieving Standard Three).

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 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 March and April 2009, to determine the proportion of vegetation production that had been consumed or destroyed by the animals. To map the use patterns, BLM drove along roads on the northern portion of the Monument and stopped every half mile to a mile to 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 that were "unsuitable" for grazing because they were too steep, and areas that were "unsurveyed." For areas with use classified as "heavy" or "severe," BLM also conducted utilization transects to verify the accuracy of the classifications.

The final use pattern map produced by BLM changed the designation of certain areas from the designation that was used on the initial use pattern map. For example, areas that were initially designated as "unsuitable" were changed to a designation of "unsurveyed or inaccessible" on the final map, and areas that were initially designated as "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.)

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 utilization (classified by BLM as negligible to slight livestock use), BLM assumed that livestock grazing was not the causal factor for non-achievement of the health standards. In areas where BLM determined that current grazing was not the causal factor for non-achievement, BLM attributed the failure to other factors such as historic livestock grazing, historic livestock patterns, fire, drought, off-road vehicle use, or general recreation.

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

Standard of Review

WWP brings this challenge under NEPA, which requires federal agencies to consider the consequences of their actions on the environment. NEPA's mandate is "essentially procedural. . . . . It is to ensure a fully informed and well considered decision." Vermont Yankee Nuclear Power Corp. v. NRDC, 435 U.S. 519, 558 (1978). "NEPA itself does not mandate particular results, but simply prescribes the necessary process." Robertson v. Methow Valley Citizens Council, 490 U.S. 332, 350 (1989). The 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 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 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 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 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 have considered the relevant information and provided a satisfactory explanation for its actions, drawing a "rational connection between the facts found and the choice made." Burlington Truck Lines v. United States, 371 U.S. 156, 168 (1962). "An agency's decision is arbitrary and capricious if it fails to consider important aspects of the issue before it, if it supports its decisions with explanations contrary to the evidence, or if its decision is either inherently implausible or contrary to governing law." In Defense of Animals, Dreamcatcher Wild Horse and Burro Sanctuary v. United States Dep't of Interior, 751 F.3d 1054, 1061 (9th Cir. 2014).

Discussion

WWP challenges the process used by BLM to reach its decision regarding grazing 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 grazing compatibility determination. WWP contends BLM failed to provide underlying data or explain how the data supported its conclusions; excluded relevant information and provided inconsistent or inadequate explanations for the exclusion; and failed to provide a rational connection between the facts in the record and BLM's conclusions. WWP challenges three steps in the LHE process: (1) the identification of the desired plant community objectives, (2) the determination of whether the identified objectives are being met, and (3) the determination of whether livestock grazing caused the failure to 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 plant community objectives. BLM contends that it did provide an adequate explanation for how it set the desired plant community objectives. The pages of the administrative record cited by BLM explain that it set objectives for each ecological site based on the average value of data from the BGR/A, or at the low end of the ESD (see AR 74601, 74608, 73126). However, as discussed below, BLM did not provide an adequate explanation in the record for adjustments it made to the BGR/A and the ESD data in setting the objectives.

1. Ecological Sites based on BGR/A Data

BLM had BGR/A data for four of the ecological sites: the limy fan, sandy wash, 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 objectives for the various ecological sites demonstrates that, although the granitic hills final objectives for canopy cover and composition are consistent with the average BGR/A data, the final objectives for the limy fan, sandy wash, and limy upland deep ecological sites are not consistent with the average BGR/A data. BLM provides an explanation in its brief for the variance between the objectives and the BGR/A data, but this explanation is not contained in the administrative record3 and, further, does not adequately explain or account for the wide variance in the adjustments made by BLM.

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, BLM explains that it further adjusted the ESD objectives for the lower precipitation north of I-8. This "lower precipitation" explanation is not, however, contained in the administrative record and, further, does not adequately explain or account for the wide variance in the adjustments made by BLM.

In addition, the LHE contains objectives for "palatable browse" for the limy 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 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.

3. Saguaro Recruitment

BLM set final saguaro recruitment objectives for only two ecological sites — the limy upland site and the granitic hills site — and based those objectives on the saguaro recruitment rate in the BGR/A. BLM explains in its brief that in setting the objectives, it adjusted downward from the BGR/A data due to lower precipitation or other variance in the ecological sites north of I-8. This explanation is not, however, contained in the administrative record and, further, does not explain the wide variance in the adjustments made.

4. Conclusion

In sum, BLM did not provide an adequate explanation in the record to support its setting of, and/or adjustments to, the objectives for the limy fan, sandy wash, limy upland deep, and granitic hills ecological sites. BLM's setting of these plant community objectives was therefore arbitrary and capricious. See Humane Soc'y of U.S. v. Locke, 626 F.3d 1040, 1049 (9th Cir. 2010) ("We cannot gloss over the absence of a cogent explanation by the agency by relying on the post hoc rationalizations offered by defendants in their appellate briefs."); Northwest Coalition for Alternatives to Pesticides (NCAP) v. U.S. E.P.A., 544 F.3d 1043, 1052 n.7 (9th Cir. 2008) ("where the agency's reasoning is irrational, unclear, or not supported by the data it purports to interpret, we must disapprove the agency's action").

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 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 and either failed to justify the omission or provided contradictory and inconsistent explanations for the omission; and failed to adequately explain or justify the use of an "80% threshold" and a "preponderance of the evidence" standard.

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 ecological sites on the allotments north of I-8 met the objectives. However, in the final LHE, BLM removed all prior data and used only the most recent 2009 data from the allotments, except for the Beloat allotment, for which BLM used 20044 data, even though it had 2007 and 2009 data for the allotment.

BLM explained that it used the best data available for its analysis, but did not 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 available data" over the more recent data. Because BLM has not provided an adequate explanation for its use of the 2004 data, rather than more recent data, as an internal BLM reviewer noted, this gives the appearance that BLM was "cherry-picking" data (AR 83853).

2. Exclusion of PBI data

BLM admits that it used some of the PBI data for its analysis, and excluded other PBI data, but contends that where it declined to use PBI data, it provided a reasonable 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 livestock water and other congregation areas, which were the most heavily impacted areas. BLM explained that it used existing methods to measure the impacts of grazing in "key areas." (AR 3857, 74357, 74612, 55414, 78494). In particular, "key areas" were "selected in locations that represent where livestock grazing pressure is occurring" and are located "approximately 1 mile (+/- 0.5 miles) from a water source to prevent 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 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.

The Court finds that BLM provided an adequate explanation for its exclusion of the PBI data from areas less than a mile from livestock water and other congregation areas, and, further, that BLM took the requisite "hard look" at, and provided a "reasonably thorough discussion of," these most heavily impacted areas. See Okanogan Highlands Alliance v. Williams, 236 F.3d 468, 473 (9th Cir. 2000); see also Northern Plains Resource Council, 668 F.3d at 1085 (NEPA requires agency to "carefully consider information about significant environmental impacts").

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 precipitation patterns. (AR 74108, 74348-49). However, a review of the record demonstrates that BLM's own plot data did not include such information, and BLM did not incorporate that information into its LHE analysis. (See AR 55485-96, 74666-82). Although the LHE report contained data about actual livestock use and precipitation, that data was not used to determine whether ecological sites met objectives. (AR 74616-18, 74622-43.) Instead, the determination of whether ecological sites met objectives was simply based on comparing the plot data to the objectives for plant cover, composition, and saguaro recruitment, and information on livestock intensity, frequency, timing, season of use, or precipitation was neither incorporated into nor considered in this 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 provide a reasonable explanation as to why it rejected some PBI data based on failure to incorporate factors such as livestock intensity, frequency, timing, season of use, or precipitation pattern factors, but then used other PBI data even though that data also did not incorporate the same factors. Absent such explanation, BLM's decision to exclude the PBI data was arbitrary and capricious. See In Defense of Animals, 751 F.3d at 1061; NCAP, 544 F.3d at 1052 n.7.

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

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 only a single year's data in its key area analysis, either from 2004 or 2009. Thus, although BLM stated that a single year's data is not enough to support sound conclusions, it chose to rely on just a single year's data for its analysis.

BLM does not provide any explanation for why it was acceptable to use just a single year's data for its determination of whether the objectives were being met, but then reject the PBI data in part because a single year's data is not enough to support sound conclusions. This lack of any reasonable explanation for the inconsistent treatment of 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. Use of Eighty Percent Threshold Approach

When comparing the 2009 or 2004 plot data to the desired plant community objectives, BLM considered the plot to be meeting an objective if the data value was within 80% of the objective (the "80% threshold"). BLM explained that it used the 80% threshold to account for site variability when determining whether an objective was being met in an individual plot (AR 74621). Thus, if a key area attribute, such as canopy cover, was within 80% of the objective for that ecological site, the objective was considered to have been met for that plot. (Id.)

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% 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 BGR/A data resulted in BLM "doubly accounting for site variability."

As to the source of the 80% value, BLM did not explain in the LHE why it chose that value (AR 74620), but in its response to comments, BLM stated that peer reviewers had suggested using a range around the absolute value rather than the absolute value itself to judge achievement of objectives to better represent real conditions on the ground, and that the 80% figure was based on best professional judgment (AR 74336). In addition, in response to protests, BLM stated:

(AR 78491.)

As to the contention that BLM was doubly accounting for site variability, as BLM 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 not adequately account for site variability. (See AR 55401, 55410, 55416, 83941.) To address this concern, BLM added the 80% threshold to account for site variability. (AR 74621 n.**.)

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

BLM used a preponderance of the evidence approach to determine whether objectives were achieved within an ecological site on an allotment. Under this approach, if more than half of the plots within a given ecological site on an allotment were achieving all of the plant community objectives, the ecological site was determined to be 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, 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 ecological site was deemed to be meeting land health standards.

However, BLM explained that it used the preponderance of the evidence approach because there were not a statistically significant number of plots within each ecological site (AR 74621). In addition, a peer reviewer explained why such an approach was needed:

(AR 55409-10.)

BLM's use of the preponderance approach, in light of its explanation and the peer reviewer's comment, is not "so implausible that it could not be ascribed to a difference in view or the product of agency expertise." WildEarth Guardians v. U.S. E.P.A., 759 F.3d 1064, 1070 (9th Cir. 2014).

C. Determination of Whether Grazing Caused Failure to Meet Objectives — the Causality Determination

Utilizing the 80% threshold and the preponderance of the evidence approach, BLM concluded that 127,550 acres, or 50.5% of the Monument lands north of I-8, were not achieving the land health standards and thus not achieving Standard Three. The next step for BLM was to determine whether the failure to achieve the land health standards was caused by livestock use. To make this determination, BLM used two methodologies — 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 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 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

BLM collected utilization data in 2009 on four of the six allotments north of I-8 to determine livestock use levels during the 2008-2009 grazing season. Precipitation for 2008 was slightly above average and there was moderate ephemeral production (400 to 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 using only data collected in 2009 was insufficient for the causality determination, and that using only that single year's data cannot account for the long-term effects to vegetation, or use patterns, that might occur in non-ephemeral years when livestock are grazing more perennial plants.

BLM contends that its use of a single year's data to determine whether a site was 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 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 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 not respond to the peer reviewer's concern that utilization data from only one year was relied on by BLM. In other words, the fact that utilization data measures a single year's utilization does not address peer reviewer's concern that BLM did not consider utilization 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 utilized by the livestock — did not account for the long-term effects to vegetation, or use patterns, that might occur in non-ephemeral years in which livestock graze on more perennial plants.

Finally, as discussed previously, BLM stated it could not rely on PBI study data in part because "one year of PBI data, in itself is not enough to support sound conclusions." (AR 74107.) Yet, BLM relied on one year of its own data to support its conclusion of whether livestock use was the cause of the non-achievement of land health standards. 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 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 determination arbitrary and capricious. See Western Watersheds Project v. Kraayenbrink, 632 F.3d 472, 492-93 (9th Cir. 2011) (failure to assess, consider, and respond to serious public comments can render agency decision arbitrary and capricious); see also In Defense of Animals, 751 F.3d at 1061.

2. Use Pattern Mapping

The second method used by BLM to make its causation determination was to conduct use pattern mapping on the Monument in March and April 2009. This method 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 "not surveyed," while unsuitable areas were marked as "unsuitable," and areas that did not have key forage species were marked as "no forage species." In the final version of the use pattern map, BLM made many changes to the classifications. Areas that had been designated as "unsurveyed" or "no forage species" on the original map were changed to the designation of "negligible" use or "slight" use in the final version of the map, and high elevation areas that were designated as "unsuitable" on the original map had been changed to "unsurveyed or inaccessible."

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 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 account in response to protest); AR 73490 ("In the absence of quantitative data, best professional judgment is used"); AR 74649 (map showing key areas and PBI plots in wilderness areas and water sources); AR 74650 (use pattern map showing water sources). BLM also noted that livestock use is slight to negligible in steep, rocky terrain, far from available water, which is characteristic of wilderness areas (AR 74156). Thus, BLM contends that, based on its professional judgment and quantitative utilization measurements, it reasonably estimated use of livestock in wilderness areas in later drafts and in the final utilization map. BLM also cites to language located at AR 78529, which states:

(AR 78529.)

The Court has reviewed the citations to the record provided by BLM, as well as the pattern use maps, and concludes that the evidence in the record provides an adequate explanation for the changes to the map, and that, accordingly, BLM's final use pattern map, and reliance on that map, was reasonable.

Conclusion

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 that this is one of those "rare circumstances" in which remand without vacatur is appropriate. See Humane Society of United States v. Locke, 626 F.3d 1040, 1053 & n.7 (9th Cir. 2010); Idaho Farm Bureau Fed'n v. Babbitt, 58 F.3d 1392, 1405-06 (9th Cir. 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 cure a defect in its explanation of a decision"). The Court will therefore grant summary judgment in favor of WWP to the extent it seeks remand to the BLM, but will deny summary judgment, without prejudice to renewal, to the extent WWP seeks to vacate BLM's decision. The Court will deny BLM's motion for summary judgment without prejudice to renewal. The Court will remand to BLM to provide it with the opportunity to either articulate reasoned explanations and responses, or adopt different decisions with reasoned explanations that support them.

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.

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.

IT IS FURTHER ORDERED that Bureau of Land Management shall file a supplemental report, providing the required reasoned explanations and responses, or indicating that it will be adopting different decisions with reasoned explanations that 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, 401 W. Washington Street, Phoenix, Arizona 85003.

IT IS FURTHER ORDERED that the parties shall file a joint status report at least fourteen days before the Status Hearing discussing the status of the case and how this case should proceed.