facts found and the choices made.’” Id. Moreover, we must be particularly careful

to avoid thinking of ourselves as “‘a panel of scientists’” rather than showing the

deference owed to the agency’s scientific specialists. Id.

(A) NEPA2

NEPA required that BLM “‘consider every significant aspect of the

environmental impact’” of the Project3 and inform the public that it had done so.4

In carrying out those functions, it prepared the EA and determined that it need not

prepare an Environmental Impact Statement. See Native Ecosystems Council v.

U.S. Forest Serv. (Native Ecosystems Council), 428 F.3d 1233, 1238–39 (9th Cir.

2005). It, therefore, issued a finding to that effect. See id. at 1239. Klamath-

2 National Environmental Policy Act of 1969, 42 U.S.C. §§ 4321–4370h. 3 Conservation Cong. v. Finley, 774 F.3d 611, 615 (9th Cir. 2014). 4 Id.

3 Siskiyou asserts that BLM did not take the required “hard look” before it reached

its decision,5 but we disagree.

We have carefully reviewed the administrative record and agree with the

district court that the EA was sufficient and justified BLM’s actions. Succinctly

put:

(1) BLM was not required to set forth the exact locations and

numbers of trees that would, or might, be removed for the purpose of exerting

some control over mistletoe growth without ignoring that parasite’s (disease’s)

beneficial effects in the forest ecosystem. It was sufficient for the EA to describe

what forest conditions then existed and to carefully explain what, by using the

proposed management practices, the forest conditions would be after BLM had

done the proposed density and disease control work.

(2) The EA and BLM decision did not take what amounted to

antithetical positions regarding old-growth trees. BLM proposed to remove some

mistletoe-infected trees and to leave old-growth trees. True, the latter might

have had mistletoe infections. But BLM made it clear that old-growth trees, as

defined in the EA, would not be removed, and nothing in the record indicates the

5 See Ctr. for Biological Diversity v. Salazar, 695 F.3d 893, 916–17 (9th Cir. 2012); Native Ecosystems Council, 428 F.3d at 1239.

4 contrary. Put otherwise, the fact that some mistletoe-infected trees would be

removed and that some old-growth trees were infected does not lead to a logical

conclusion that some old-growth trees would be removed.

(3) Klamath-Siskiyou asserts that improper (illegal) off-highway-

vehicle incursions might occur. But the EA recognized that possibility and

carefully detailed steps that would be taken to preclude, or at least greatly

minimize, the problem. BLM reasonably concluded that there would be few or no

impacts and sufficiently discussed the risks posed by off-highway-vehicle use in

light of that conclusion. Of course, there is and always will be the possibility of

illegal activity by others, but if that were sufficient to preclude BLM action, it

could never act at all.

(B) FLPMA6

Under FLPMA, BLM was required to propose a proper land use plan,7 and

once that was adopted, BLM was required to adhere to it.8 Our review of the

6 Federal Land Policy and Management Act of 1976, 43 U.S.C. §§ 1701- 1787. 7 See Klamath Siskiyou Wildlands Ctr. v. Boody, 468 F.3d 549, 555–56 (9th Cir. 2006). Here that plan was the Medford District Resource Management Plan (“the Plan”). 8 See Or. Nat. Res. Council Fund v. Brong, 492 F.3d 1120, 1125 (9th Cir. (continued...)

5 record satisfies us that BLM did adhere to the Plan. Again, succinctly put:

(1) Klamath-Siskiyou maintains that the Plan required that every

management action by BLM maintain or improve soil productivity and that the

Project will decrease productivity to some extent. However, that misstates the

Plan, which, rather, states that soil improvement is a desideratum and requires use

of the best management practices, but which also recognizes that any surface

disturbance may cause some relatively minor losses, even though many steps will

be taken to minimize those while improving the overall health of the forest. In

short, there is no absolute maintain-or-improve rule. Therefore, the losses of soil

productivity that are expected to result from the Project are permitted by the Plan.

The district court did not err.

(2) Klamath-Siskiyou then asserts that the EA is not consistent with

the Plan’s provisions regarding fragile soils. Klamath-Siskiyou did not specifically

raise its fragile-soils claim during the administrative proceedings, as it should have

done. See Lands Council, 629 F.3d at 1076. Nevertheless, we will consider the

argument on its merits. See id.; cf. Buckingham v. Sec’y of the U.S. Dep’t of

Agric., 603 F.3d 1073, 1080–81 (9th Cir. 2010). Again, however, the district court

8 (...continued) 2007); Klamath Siskiyou, 468 F.3d at 556.

6 did not err. The record shows that in the EA BLM did carefully consider the

nature of the soils that it would impact within the Project area, and it did not find

any of them to be fragile. Unsurprisingly, Klamath-Siskiyou did not point to any

evidence in the record that there were fragile soils; it did not refer to fragility at all

during the administrative proceedings. Of course, Klamath-Siskiyou does now

claim that Map 6 in the Plan is sufficient to show that fragile soils will be

impacted. We disagree. That map is, as BLM pointed out, an indication of areas

that might contain fragile soils rather than a determination that those areas do

contain fragile soils, in whole or in any particular part. That is something that had

to be determined on the ground when a project was proposed.

The disclaimer on Map 6 is expansive: “The Bureau of Land Management

cannot assure the reliability or suitability of this information for a particular

purpose. Original data was compiled from various sources. Spatial information

may not meet National Map Accuracy Standards. This information may be

updated without notification.” We find it counterintuitive that Map 6 details

definitive fragile-soils designations in the face of clear language disclaiming “the

reliability or suitability of this information for a particular purpose” and warning

that the “information may be updated without notification.” Instead, we hold that

Map 6 is not a binding determination of the exact locations of fragile soils.

7 The dissent points out that the same disclaimer is on all but one of the other

maps in the RMP. This fact does not change our conclusion. Like Map 6, all the

other maps with the disclaimer are illustrative rather than depictions creating static,

binding obligations on BLM. For instance, many of the other maps detail land

designations that have changed since the RMP’s creation in 1995. To hold that the

maps from 1995 require BLM to manage the land in conformity with outdated land

designations would be illogical.

Finally, both the dissent and Klamath-Siskiyou claim that our holding

hollows out BLM’s obligations under the RMP because BLM can now avoid those

obligations by simply changing geographic designations without public oversight.

We respectfully disagree. Our holding does not grant BLM the carte blanche right

to trample on fragile soils. BLM is undisputedly bound by the RMP, including its

fragile-soils requirements. We simply hold that Map 6 is not an exact

representation of where those fragile soils actually exist. Accordingly, the

fragility of the soil must be determined from a project-specific soil analysis rather

than from a coarse-scale map covering 859,000 acres.

AFFIRMED.

8

FILED

FEB 29 2016

Klamath-Siskiyou Wildlands Ctr. v. Gerritsma, No. 13-35811 MOLLY C. DWYER, CLERK

U.S. COURT OF APPEALS

Bea, Circuit Judge, Concurring and Dissenting I join my colleagues’ analysis as to Plaintiffs’ NEPA claims in full. I write separately, however, because I would find that Plaintiffs have indeed demonstrated that the Bureau of Land Management’s (“BLM’s”) authorization of the Rio Climax Project (the “Project”) violated the Medford Resource Management Plan’s (the “RMP’s”) “fragile soil” provisions. The Federal Land Policy and Management Act (“FLPMA”) requires the BLM to adopt a land use plan—here, the Medford RMP—which imposes substantive limitations on the BLM’s discretion. See 43 U.S.C. § 1732(a); 43 C.F.R. §1610.5-3(a). The Medford RMP, as approved by the Secretary of the Interior, includes various Appendices and thirteen maps, including Map 6. For the reasons set forth below, I read Map 6 as memorializing binding determinations regarding the location of “sensitive,” or fragile, soils in the Medford area. Thus, the Project’s authorization of tractor logging over designated fragile soil areas violates the “best management practices” mandated in the RMP’s Appendix D (instructing the BLM to use suspension logging and to “avoid tractor logging” over fragile soils). The BLM’s failure to comply with its governing RMP, in turn, violates the FLMPA. Independently, the BLM’s attempt to alter its RMP (by reclassifying soils designated as fragile in Map 6) without complying with statutorily mandated amendment procedures violates the FLPMA. For both of these reasons, I would reverse the district court’s grant of summary judgment to the BLM on Plaintiffs’ FLPMA claim and enter judgment in favor of Plaintiffs on

-1- that claim.1 See Oregon Nat. Res. Council Fund v. Brong, 492 F.3d 1120, 1125–31 (9th Cir. 2007) (finding that the BLM’s authorization of a project based on an interpretation of its RMP that was “plainly inconsistent” with that RMP violated the FLPMA, and entering judgment in favor of Plaintiffs on that claim). In holding that Map 6 is not a binding determination of the location of fragile soils, the majority relies exclusively on the disclaimer language written in illegibly small text on Map 6, which states: BLM cannot assure the reliability or suitability of this information for a particular purpose. Original data was compiled from various sources. Spatial information may not meet National Map Accuracy Standards. This information may be updated without notification. ER 152. The BLM and the majority read far too much into boilerplate language developed by the United States Geological Survey (“USGS”) that commonly appears on maps produced by the BLM—as well as on maps produced by other

1 As a threshold matter, I agree with the majority’s conclusion that Plaintiffs have not waived this argument by failing specifically to mention the term “fragile soil” during the administrative proceedings and instead focusing more broadly on their concerns of soil degradation and erosion. We have held that plaintiffs can make more specific arguments in court proceedings than they made in administrative proceedings, so long as the plaintiff previously raised the broader issue and thus gave the agency adequate notice of it. See Native Ecosystems Council v. Dombeck, 304 F.3d 886, 899–900 (9th Cir. 2002). Here, the RMP’s fragile soil and soil productivity provisions appear in the same three-page section regarding soil directives. Moreover, they work in tandem with each other. (See, e.g., “[S]ome minor losses in productivity could result due to surface disturbances (soil compaction, road construction, etc.) caused by management activities. Implementing best management practices and minimizing disturbance of fragile areas will keep losses to a minimum (see Appendix D).” (emphases added)). Thus, I would find that Plaintiffs adequately preserved their current fragile soil argument.

-2- departments of our federal government.2 See National Map Accuracy Standards, U.S. GEOLOGICAL SURVEY (rev. June 17, 1947) (last visited Feb. 22, 2016), available at http://nationalmap.gov/standards/nmas.html. In my view, the disclaimer on Map 6 is nothing more than a warning to the public to do its own investigation, appropriate to its own purposes. To start, the majority’s interpretation is inconsistent with the text of the disclaimer itself: That the BLM is the speaker in the disclaimer (i.e. “BLM cannot assure . . .”) strongly suggests that the disclaimer is directed to persons other than the BLM who might see the maps in the RMP. In this regard, the disclaimer is similar to warnings with which investors are familiar: In the preparation of this site, every effort has been made to offer the most current, correct and clearly expressed information possible. Nonetheless, inadvertent errors can occur . . . . Users are encouraged to consult with professional advisers for advice concerning specific matters before making any decision impacting on these matters.” Disclaimer, PI TRADING.COM (last visited Feb. 23, 2016), available at http://pitrading.com/disclaimer.html. Such disclaimers are pervasive in our society. Yet one does not often find financial advisors stating they themselves do not rely on the information upon which they themselves have developed. The interpretation of Map 6’s disclaimer adopted by the majority today—that the BLM wrote the disclaimer on Map 6 to itself—defies common sense. If the BLM is telling itself, and everyone else, not to rely on the maps it draws up, of what earthly use is a map of fragile soils? (Pardon the pun). Indeed, the very purpose of land use plans, of which maps are a part, is to limit the 2 See, e.g., Chugach National Forest GIS – Ownership Boundaries, U.S. DEP’T OF AGRIC. (Nov. 29, 2011), available at http://data.fs.usda.gov/geodata/rastergateway/alaska/chugach/landstat.php.

-3- discretion of the BLM in managing natural resources. If the maps are not to be followed, how can they limit the BLM’s discretion? The majority’s interpretation also controverts the plain text and structure of the RMP pertaining to fragile soils, which directs the BLM: [To] [m]anage lands dominated by fragile granitic and schist soils consistent with southern general forest management area guidelines [and to] . . . limit surface-disturbing activities on all lands dominated by fragile granatic, schist, and pyroclastic soils (approximately 85,300 acres) to maintain site productivity, reduce soil erosion, and minimize water quality degradation. These soils are scattered throughout the planning area, however, the largest concentrations of soils formed from decomposed schist and/or granite parent material occurs [sic] in Evans Snow, Sugar, and Meadow Creeks, upper portions of Williams Creek, and headwaters of Birdseye Creek. Soils formed in deeply weathered pyroclastic parent materials are predominantly in the foothills of the Cascades. (See Map 6 and Appendix D for fragile soils mitigation measures.) (emphases added). The majority’s conclusion that Map 6 is illustrative only is inconsistent with this passage’s use of unqualified, affirmative statements in describing the location and existence of fragile soils in the project area. The RMP does not say that the BLM “thinks” or “believes” there may be fragile soils in the project area. Rather, the RMP declares that fragile soils “are scattered throughout the planning area.” (emphasis added). It specifically identifies and describes regions where “the largest concentrations of soils . . . occur[].” It says flatly that “deeply weathered pyroclastic parent materials are predominantly in the foothills of the Cascades.” (emphasis added). It even quantifies the total acreage containing fragile soils (85,300 acres) with relative specificity. And after generally describing

-4- the regions containing fragile soils, the RMP directs the reader to “Map 6” for more details.3 Furthermore, the BLM’s argument that the disclaimer renders Map 6 merely illustrative is inconsistent with the presence of the same disclaimer on every map in the Medford RMP’s “map packet” (with the exception of Map 1).4 If the BLM is not bound by Map 6’s fragile soils designation, then it is also free to disregard the RMP’s watershed designations (Map 5), its big game designations (Map 7), its “special area” designations (Map 8), and all the substantive provisions that apply to those areas.5 Such a result cannot be squared with the Medford RMP’s routine reliance on the map packet to identify the locations of certain resources and landscapes where the RMP’s directives and best management practices apply. Indeed, the descriptions in the RMP’s main text of the lands to which its mandates apply are often so general and cryptic that one would find it difficult, if not impossible, to determine specifically where those mandates apply without the maps. Like the soil

3 That Map 6 should be interpreted as binding on BLM is further corroborated by the RMP’s introductory statement that “[m]aps of the resource management plan land use allocations are located in the accompanying map packet.” (emphasis added). In other words, the thirteen maps incorporated into the Medford RMP are not superfluous; their purpose is to make “land use allocations” in the Medford area. 4 The parties have not included every map in their excerpts of record, but the Medford RMP’s map packet is publicly available at http://www.blm.gov/or/plans/files/RMP_1995_Medford_Maps.pdf. These maps are therefore subject to judicial notice. 5 That is, BLM could simply avoid any limitations associated with those geographical designations by unilaterally reclassifying the underlying land designation shown in the map packet—without any public or agency oversight.

-5- directives section corresponding to Map 6, quoted above, the main RMP section pertaining to “Riparian Reserves” sets forth various management directives that apply in “key watershed[]” areas, but gives virtually no information about where those watersheds are located. Concededly, the RMP lists the “Key Watershed Name[s]” and the number of “BLM Acres” associated with each watershed (e.g., “2,710” acres are classified as “Beaver Creek”), but this information alone is clearly insufficient to determine precisely how those 2,710 acres designated as the “Beaver Creek” watershed are distributed throughout the project area and, in turn, where the RMP’s watershed provisions apply. The RMP therefore directs the reader to “[s]ee Map 3 for locations of key watersheds.” Similarly, the RMP section corresponding to Map 8 describes management directives pertaining to “areas of critical environmental concern,” and repeatedly directs the reader to “Map 8 for locations and Table 6 for site-specific acres.” See also (“Actions including timber harvest will be allowed if they do not conflict with the habitat needs of these plants (see Map 8)). Management 16,340 acres near Soda Mountain and Agate Flat areas as the Cascade/Siskiyou ecological emphasis area (see Map 8). . . . Three areas, Ferris Gulch (2,200 acres), Timber Mountain/John’s Peak (16,250 acres), and Quartz Creek (7,120 acres) will be managed to provide for OHV use. See Table 8.”). In sum, it is clear from these passages that the purpose of the map packet is to identify the land areas over which the RMP’s numerous directives apply. The majority’s holding that the BLM is bound only by the RMP’s directives—but not by a map packet whose clear

-6- purpose is to demonstrate the locations where those directives apply—is fundamentally inconsistent with the structure and text of the RMP.6 My colleagues urge that the maps are merely illustrative because land designations may change over time (and have changed since the approval of the Medford RMP in 1995). Thus, to require the BLM to manage lands in conformity with outdated land designations would be “illogical.” There are three problems with the majority’s logic. First, the same argument could be made as to the land use designations listed in the RMP itself. But the BLM does not dispute that those designations are binding.7 Second, the majority’s argument is inconsistent with the RMP itself, as well as basic canons of statutory interpretation. The RMP does not require the BLM to comply with static land designations; on the contrary, it vests the BLM with significant discretion to alter some land use designations without complying with formal FLPMA amendment procedures.8 However, it confers no comparable discretion with respect to fragile 6 Indeed, the BLM’s arguments are internally inconsistent. On the one hand, the BLM argues that it is not bound by the map packet because (as demonstrated by the disclaimer language) the maps are too imprecise. Yet the BLM concedes that it is bound by the less precise land use designations in the RMP. 7 For example, the main RMP text routinely quantifies the specific acreage associated with various land use designations, even though that acreage would necessarily change if, as the majority suggests, the land use designations shown on in the map packet change. (See, e.g., Table 6, listing the acreage and management prescriptions for various “special” management areas). 8 For example, the RMP permits the BLM to modify the boundaries of the north and south “general forest management areas” if the BLM “determines that portion of the area are no longer within ½-mile of raptor nests which have been active within the past two years.” Likewise, the RMP authorizes the BLM to modify the area designated as “crucial winter range for wildlife [primarily migrating elk and deer]” if the BLM “determines that portions of the area no longer contain crucial

-7- soil designations.9 The principle of expressio unius requires us to infer from this omission that the BLM does not have authority to modify fragile soil designations through a short-cut process that evades public oversight. Which brings me to the majority’s third error: its assumption that a simple finding that the whole RMP (i.e., the RMP as approved by the Secretary of the Interior) is binding would require the BLM to comply with outdated designations. It would not. It would simply require the BLM to comply with well-established FLPMA amendment procedures—which exist to address the very situation that (according to the BLM) has arisen here.10 If in fact there are no fragile soils in the project area, then the BLM will be able successfully to modify its RMP through this statutorily mandated procedure. But the proper solution for correcting an inaccurate map is for the BLM to propose an amendment and submit an Environmental Assessment for public review and comment—and not for us to

winter range for wildlife.” 9 The BLM argues that the RMP confers on the BLM discretion to “designate new RNAs [‘research natural areas’] as appropriate.” But this provision merely permits the BLM to increase restrictions on its land use. It hardly follows that the RMP gives the BLM equal autonomy to remove land use restrictions. 10 The BLM relatedly argues that minor discrepancies between the maps and the main body of the RMP show that the map packet must be advisory (for example, Map 9 depicts more “recreation sites” than are listed in the corresponding RMP provision). Putting aside that this is actually consistent with the structure of the RMP (under which the maps often provide far more detailed and precise information than the main text of the RMP), this argument fails for the same reason. To the extent the BLM thinks that any maps are inaccurate (or no longer accurate), then—absent a provision in the RMP to the contrary—BLM needs to comply with statutorily prescribed amendment procedures to make those maps accurate. Until then, the BLM is bound by the land use designations approved by the Secretary of the Interior.

-8- adopt a textually, logically, and functionally unsupportable reading of the BLM’s land use plan. In fact, our precedent makes clear that the BLM’s attempt to reclassify previously designated fragile soil areas without following formal amendment procedures violates the FLPMA. See Klamath Siskiyou Wildlands Center v. Boody, 468 F.3d 549 (9th Cir. 2006).11 As in Boody, BLM’s 2011 soil analysis effectively amends the RMP’s findings relating to the location of fragile soils. By finding that previously designated “fragile soil” areas are no longer fragile, BLM altered the “terms and conditions” of the RMP by eliminating the necessity of

11 In Boody, we held that agency actions that effectively “amend . . . resource management plans” must comply with formal NEPA and FLPMA procedures. Id. at 560. There, the BLM’s original land management plan assigned the red tree vole a “Category C” protection rating, which meant that certain measures had to be taken before the BLM could authorize any actions that would disturb the red vole’s habitat. Id at 553. Then, without following NEPA or FLPMA procedures, the BLM issued a memorandum unilaterally downgrading the red tree vole to “Category D,” and then later eliminated protections altogether. Id. These downgrades were done in connection with a timber sale that would almost certainly “destroy” any red vole nests in the harvest area. Id. By downgrading the red vole’s protected status, the BLM avoided having to follow the protection procedures prescribed by the governing land management plan. Id. We held that changing the protection status assigned to the red vole in the management plan qualified as an “amendment” (as opposed to mere plan “maintenance”) because it “altered the terms and conditions of the [management plan].” Id. at 556-57; see 43 C.F.R. § 1610.5–5 (summarizing amendment procedures). We therefore concluded that the BLM violated both NEPA and the FLPMA by changing a term or condition of a management plan without following proper statutory procedures. Boody, 468 F.3d at 559-62. Furthermore, we concluded, the timber sales authorized on the basis of that unauthorized plan change also violated federal law. Id.at 563.

-9- complying with best management practices in those areas. Under Boody, this unauthorized amendment violates both NEPA and the FLPMA. BLM argues that even if Map 6 is binding, the RMP requires only that the BLM “avoid” tractor yarding over fragile soils; but it does not strictly “prohibit” such logging. The majority does not reach this argument, so I address it only briefly. The BLM’s apparent suggestion that an RMP’s proscriptions must be absolute before the BLM need comply with them is untenable. The command to “stop” on the red placards dotting our streets is not optional simply because those placards do not read “stop absolutely.” Similarly here, Merriam Webster’s defines the term “avoid” as “to prevent the occurrence,” or “to refrain from [an act].” MERRIAM-WEBSTER 3b-3c (last visited Nov. 23, 2015). Thus, the RMP’s instruction to “avoid” tractor logging means the BLM must “refrain” from or “prevent the occurrence” of tractor logging over fragile soils—an instruction that the BLM clearly violates in affirmatively authorizing tractor logging over those areas. In any event, the argument that the BLM can ignore the RMP’s restrictions simply because they are not stated in absolute terms proves too much. All provisions of the RMP are framed in similarly aspirational terms. The interpretation urged by the BLM would therefore render illusory all of the RMP’s terms.12

12 I also note that the BLM has not even attempted to show that use of “best management practices,” including “full or partial suspension . . . yarding” in order to avoid surface-disturbing activity over fragile soils identified in Map 6, would be impracticable here. Suspension yarding utilizes a suspended pulley system whereby harvested logs are transported above the ground on a suspended steel cable (called a “highline”) from locations where trees are felled to a central location (usually a road head) for further transport. Suspension yarding reduces soil disturbances because the logs do not drag along the ground.

-10- For the foregoing reasons, I would reverse the district court’s grant of summary judgment to the BLM on Plaintiffs’ FLPMA claim. To the extent the majority concludes otherwise, I respectfully dissent.

-11-