2

3

UNITED STATES DISTRICT COURT

4

FOR THE EASTERN DISTRICT OF CALIFORNIA

5

6 CENTRAL SIERRA ENVIRONMENTAL 1:17-cv-00441-LJO-SAB RESOURCE CENTER, et al., 7 MEMORANDUM DECISION AND Plaintiffs, ORDER RE CROSS-MOTIONS FOR 8 SUMMARY JUDGMENT v. 9 (ECF Nos. 80, 85, and 86) STANISLAUS NATIONAL FOREST, et al., 10

11 Defendants.

12

13 I. INTRODUCTION

14 Plaintiffs Central Sierra Environmental Resource Center (“CSERC”) and Sierra Forest Legacy

15 (together, “Plaintiffs”) bring this suit against Defendants Stanislaus National Forest, the U.S. Forest

16 Service (“Forest Service” or “USFS”), and Jason Kuiken, in his official capacity as Forest Supervisor

17 for the Stanislaus National Forest1 (together, “Federal Defendants”), challenging the cattle grazing

18 program for three livestock allotments in Stanislaus National Forest. ECF No. 57. Following briefing

19 and a decision on a motion to dismiss, Plaintiffs filed a Third Amended Complaint (“TAC”), bringing

20 claims under the Administrative Procedure Act premised on violations of the Clean Water Act and the

21 National Forest Management Act. Id. Plaintiffs moved for summary judgment. ECF No. 75. Federal

22 Defendants; and Defendant-Intervenors, various permittees and the permittees’ trade association filed

23 24

25 1 The Court is sep arately issuing an order substituting Jason Kuiken as a party, replacing former Forest Supervisor Jeanne M. 2 reply in a single filing. ECF No. 87. Federal Defendants, ECF No. 88, and Defendant- Intervenors,

3 ECF No. 90, both filed replies. This matter is now ripe for review and has been deemed suitable for

4 disposition without oral argument. See Local Rule 230(g); ECF No. 91.

5 II. STATUTORY BACKGROUND

6 A. Clean Water Act

7 The purpose of the Clean Water Act (“CWA”) is “to restore and maintain the chemical, physical,

8 and biological integrity of the Nation’s waters.” 33 U.S.C. § 1251. Accordingly, the CWA prohibits

9 “the discharge of any pollutant by any person” into waters of the United States except when discharged

10 in compliance with a National Pollution Discharge Elimination System (“NPDES”) permit. 33 U.S.C.

11 §§ 1311(a), 1342. The CWA “drew a distinct line between point and nonpoint pollution sources.”

12 Oregon Nat. Res. Council v. U.S. Forest Serv., 834 F.2d 842, 849 (9th Cir. 1987). The CWA defines

13 point sources as “discernible, confined and discrete conveyances,” including pipes and ditches. 33

14 U.S.C. § 1362(14).2 The CWA does not define nonpoint sources, but they consist of other sources of

15 pollution that do “not result from the ‘discharge’ or ‘addition’ of pollutants from a point source.”3

16 Oregon Nat. Res. Council, 834 F.2d at 849 n.9. Nonpoint sources of pollution include runoff from

17 animal grazing and irrigated agriculture. Oregon Nat. Desert Ass’n v. Dombeck, 172 F.3d 1092, 1095

18 (9th Cir. 1998) (“O.N.D.A. v. Dombeck”). The parties do not dispute that “something as inherently

19 mobile as a cow” represents a nonpoint pollution source. Id. at 1099.

20 The CWA directly regulates pollution from point sources through the issuance of NPDES

21 permits but “provides no direct mechanism to control nonpoint source pollution.” O.N.D.A. v.

22

23 2 The full definition is as follows: “The term ‘point source’ means any discernible, confined and discrete conveyance, including but not limited to any pipe, ditch, channel, tunnel, conduit, well, discrete fissure, container, rolling stock, 24 concentrated animal feeding operation, or vessel or other floating craft, from which pollutants are or may be discharged. This term does not include agricultural stormwater discharges and return flows from irrigated agriculture.” 33 U.S.C. § 1362(14). 25 3 The CWA defin es “discharge of pollutant” to mean, in relevant part, “any addition of any pollutant to navigable waters 2 states to accomplish this task” through federal grants for state wastewater treatment plans, 33 U.S.C.

3 § 1288(b)(2), and a requirement that states prepare nonpoint source management programs, 33 U.S.C.

4 § 1329. The latter provision, CWA § 319, “does not require states to penalize nonpoint source polluters

5 who fail to adopt best management practices; rather it provides for grants to encourage the adoption of

6 such practices.” Nat. Res. Def. Council v. E.P.A., 915 F.2d 1314, 1318 (9th Cir. 1990). California’s

7 Porter-Cologne Water Quality Control Act (“Porter-Cologne Act” or, simply, “Porter-Cologne”)

8 established California’s framework for water quality regulation in the state. Cal. Water Code (“CWC”)

9 § 13000 et seq. The Porter-Cologne Act vested California’s State Water Resources Control Board

10 (“State Water Board”), CWC § 13100, and nine regional water boards, CWC § 13200, with power to set

11 standards and procedures to protect water quality, such as the creation and adoption of water quality

12 control plans, CWC § 13240, and control over the information that waste dischargers must file with the

13 regional board, CWC § 13260.

14 Though the CWA does not itself regulate nonpoint pollution sources, it provides that federal

15 agencies are required to comply with state and local water quality requirements to the same extent as

16 nongovernmental actors. CWA § 313, 33 U.S.C. § 1323. This requirement applies both to point and

17 nonpoint sources. O.N.D.A. v. Dombeck, 172 F.3d at 1098 (“Section 1323 plainly applies to nonpoint

18 sources of pollution on federal land.”).

19 B. National Forest Management Act

20 The Forest Service manages the National Forests pursuant to the National Forest Management

21 Act of 1976 (“NFMA”). See 16 U.S.C. §§ 1600-1614. The NFMA and its implementing regulations

22 provide for forest planning and management at the forest level and at the individual project level. See

23 id.; see also Inland Empire Pub. Lands Council v. U.S. Forest Serv., 88 F.3d 754, 757 (9th Cir. 1996).

24 At the forest level, the Forest Service is required to develop a Land and Resource Management Plan

25 (“LRMP” or “Forest Plan”), which operates as a long-term planning document for an entire National 2 At the individual project level, site-specific actions, such as resource plans, contracts, and grazing

3 permits, are approved or denied by the Forest Service consistent with the governing LRMP. See Inland

4 Empire Pub. Lands Council, 88 F.3d at 757.

5 “It is well-settled that the Forest Service’s failure to comply with the provisions of a Forest Plan

6 is a violation of NFMA.” Native Ecosystems Council v. U.S. Forest Serv., 418 F.3d 953, 961 (9th Cir.

7 2005). NFMA is clear that “[r]esource plans and permits, contracts, and other instruments for the use

8 and occupancy of National Forest System lands shall be consistent with the land management plans.”

9 16 U.S.C. § 1604(i); see also Neighbors of Cuddy Mountain v. Alexander, 303 F.3d 1059, 1062 (9th Cir.

10 2002) (“Specific projects, such as [a specific] timber sale, must be analyzed by the Forest Service and

11 the analysis must show that each project is consistent with the plan.”); Idaho Sporting Cong., Inc. v.

12 Rittenhouse, 305 F.3d 957, 962 (9th Cir. 2002) (“[A]ll management activities undertaken by the Forest

13 Service must comply with the forest plan, which in turn must comply with the Forest Act.”); Neighbors

14 of Cuddy Mountain v. U.S. Forest Serv., 137 F.3d 1372, 1377–78 (9th Cir. 1998) (holding that the Forest

15 Service was not in compliance with NFMA where its site-specific project was inconsistent with the

16 forest plan of the entire forest); Friends of Southeast’s Future v. Morrison, 153 F.3d 1059, 1068 n.4 (9th

17 Cir. 1998) (“16 U.S.C. § 1604(i) plainly imposes a legal obligation on the Forest Service to ensure that

18 timber sales are consistent with the relevant Forest Plan.”).

19 The Forest Service authorizes grazing on allotments through three types of site-specific actions,

20 each of which must be consistent with the applicable Forest Plan. Buckingham v. Sec’y of U.S. Dep’t of

21 Agr., 603 F.3d 1073, 1077 (9th Cir. 2010). The first type of action is grazing permits, which are

22 “document[s] authorizing livestock to use National Forest System or other lands under Forest Service

23 control for the purpose of livestock production.” 36 C.F.R. § 222.1(b)(5); see also 43 U.S.C. §§

24 1702(p), 1752(a). Grazing permits ordinarily specify “(1) the number, (2) kind, (3) and class of

25 livestock, (4) the allotment to be grazed, and (5) the period of use.” Buckingham, 603 F.3d at 1077 2 U.S. Forest Serv.”)). The standard term for grazing permits is ten years. Id. (citing 43 U.S.C. § 1752(b);

3 36 C.F.R. § 222.3(c)(1)). The Forest Service “‘is authorized to cancel, modify, or suspend grazing and

4 livestock use permits in whole or in part’ if the permittee fails to comply with the requirements of his or

5 her permit, or with governing regulations.” Id. (quoting 36 C.F.R. § 222.4(a)(4)).

6 The second type of site-specific action is an “allotment management plant” (“AMP”), which is

7 “a document that specifies the program of action designated to reach a given set of objectives” as to a

8 specific allotment, including “the manner in and extent to which livestock operations will be conducted

9 in order to meet the multiple-use, sustained yield, economic, and other needs and objectives as

10 determined for the lands, involved.” Id. (citing 36 C.F.R. § 222.1(b)(2)). If no AMP has been

11 completed or if the Forest Service determines that none is necessary, then the grazing permits and leases

12 include “such terms and conditions as [the Forest Service] deems appropriate for management of the

13 permitted or leased lands.” 43 U.S.C. § 1752(e).

14 Finally, the third type of site-specific action is the development of annual operating plans

15 (“AOPs”) or instructions (“AOIs”). “Whereas the AMP relates the directives of the applicable [F]orest

16 [P]lan to the individual grazing allotment . . . the AOI annually conveys these more long-term directives

17 into instructions to the permittee for annual operations.” O.N.D.A. v. U.S. Forest Serv., 465 F.3d at 980.

18 “Because an AOI is issued annually, it is responsive to conditions that the Forest Service could not or

19 may not have anticipated and planned for in the AMP or grazing permit . . . .” Id. at 980-81. The terms

20 of the AOI are made part of the grazing permit, which then “governs the permit holder’s grazing

21 operations for the next year.” Id. at 980.

22 C. National Environmental Policy Act

23 The National Environmental Policy Act (“NEPA”) “is our ‘basic national charter for protection

24 of the environment.’” Ctr. for Biological Diversity v. Nat’l Highway Traffic Safety Admin., 538 F.3d

25 1172, 1185 (9th Cir. 2008) (quoting 40 C.F.R. § 1500.1). “Although NEPA does not impose any 2 Action Network v. U.S. Dept. of Transp., 545 F.3d 1147, 1153 (9th Cir. 2008). “Through these

3 procedural requirements, NEPA seeks to make certain that agencies will have available, and will

4 carefully consider, detailed information concerning significant environmental impacts, and that the

5 relevant information will be made available to the larger public audience.” Id. (internal citations and

6 quotations omitted).

7 NEPA requires federal agencies to analyze the potential environmental impacts of any “major

8 Federal actions significantly affecting the quality of the human environment.” 42 U.S.C. § 4332(2)(C).

9 When an agency takes major federal action, the agency must prepare an Environmental Impact

10 Statement (“EIS”) “where there are substantial questions about whether a project may cause significant

11 degradation of the human environment.” Native Ecosystems Council v. U.S. Forest Serv., 428 F.3d

12 1233, 1239 (9th Cir. 2005).

13 An agency may choose to prepare an environmental assessment (“EA”) to determine whether an

14 EIS is needed. 40 C.F.R. §§ 1501.4, 1508.9(b). An EA is meant to be a “concise public document . . .

15 that serves to,” among other things, “[b]riefly provide sufficient evidence and analysis for determining

16 whether to prepare an environmental impact statement or a finding of no significant impact.” 40 C.F.R.

17 § 1508.9; see also Bob Marshall Alliance v. Hodel, 852 F.2d 1223, 1225 (9th Cir. 1988). Based on the

18 EA, the agency “may conclude that the action will not significantly affect the environment and issue a

19 ‘Finding of No Significant Impact’ (‘FONSI’) in lieu of an EIS.” Bob Marshall, 852 F.2d at 1225

20 (citing 40 C.F.R. § 1508.13).

21 D. Administrative Procedure Act

22 The Administrative Procedure Act (“APA”), 5 U.S.C. §§ 701-06, provides that “[a] person

23 suffering legal wrong because of agency action, or adversely affected or aggrieved by agency action

24 within the meaning of a relevant statute, is entitled to judicial review thereof.” 5 U.S.C. § 702. Under

25 the APA, a reviewing court shall “hold unlawful and set aside agency action, findings, and conclusions 2 (A) arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law; . . . 3 (C) in excess of statutory jurisdiction, authority, or limitations, or short of 4 statutory right; [or]

5 (D) without observance of procedure required by law[.]

6 Id. § 706. When assessing claims pursuant to the APA, a court, reviewing only the AR, must determine

7 “whether or not as a matter of law the evidence in the administrative record permitted the agency to

8 make the decision it did.” Sierra Club v. Mainella, 459 F. Supp. 2d 76, 90 (D.D.C. 2006) (quoting

9 Occidental Eng’g Co. v. INS, 753 F.2d 766, 769 (9th Cir. 1985)). In other words, a court’s “review is

10 guided by whether the agency’s analysis is reasonable and offers sufficient detail to ensure that

11 environmental consequences have been fairly evaluated.” Protect Our Communities Found. v. Jewell,

12 825 F.3d 571, 582 (9th Cir. 2016) (citations and quotation marks omitted).

13 A reviewing court “must consider whether the decision was based on a consideration of the

14 relevant factors and whether there has been a clear error of judgment.” Citizens to Preserve Overton

15 Park, Inc. v. Volpe, 401 U.S. 402, 416 (1971), abrogated in part on other grounds as recognized in

16 Califano v. Sanders, 430 U.S. 99, 105 (1977). Although a court’s inquiry must be thorough, the

17 standard of review is highly deferential; the agency’s decision is “entitled to a presumption of

18 regularity,” and a court may not substitute its judgment for that of the agency. Id. at 415-16.

19 Courts should defer to the agency on matters within the agency’s expertise unless the agency

20 completely failed to address a factor that was essential to making an informed decision. Nat’l Wildlife

21 Fed’n v. Nat’l Marine Fisheries Serv., 422 F.3d 782, 798 (9th Cir. 2005). A court “may not substitute

22 its judgment for that of the agency concerning the wisdom or prudence of [the agency’s] action.” River

23 Runners for Wilderness v. Martin, 593 F.3d 1064, 1070 (9th Cir. 2010). As the Ninth Circuit explained

24 in River Runners: 25 In conducting an APA review, the court must determine whether the 2 f co leu an rd ea rn rd o rt h oe f c jh uo di gce ms em nta .”d e A . r. i z. .a n Cd a w ttlh ee t Ghe rr o w[t eh re s ’a g Aen ssc ’y n] h va . s Uc .o Sm . m Fi it st he d & a Wildlife, 273 F.3d 1229, 1243 (9th Cir. 2001). “The [agency’s] action . . . 3 need only be a reasonable, not the best or most reasonable, decision.” Nat’l Wildlife Fed’n v. Burford, 871 F.2d 849, 855 (9th Cir. 1989). 4 River Runners, 593 F.3d at 1070. Reviewing courts must be at their “most deferential” when an agency

5 makes predictions, “within its area of special expertise, at the frontiers of science.” Baltimore Gas &

6 Elec. Co. v. Nat. Res. Def. Council, 462 U.S. 87, 103 (1983). In particular, an agency’s “scientific

7 methodology is owed substantial deference.” Gifford Pinchot Task Force v. U.S. Fish & Wildlife Serv.,

8 378 F.3d 1059, 1066 (9th Cir. 2004), superseded on other grounds by regulation as stated in Defenders

9 of Wildlife v. Zinke, 856 F.3d 1248, 1260 (9th Cir. 2017).

10 But “the deference accorded an agency’s scientific or technical expertise is not unlimited.”

11 Brower v. Evans, 257 F.3d 1058, 1067 (9th Cir. 2001). Deference is not owed if “the agency has

12 completely failed to address some factor consideration of which was essential to making an informed

13 decision,” id. (internal citation and quotation omitted), and courts are not required to defer to an agency

14 conclusion that runs counter to that of other agencies or other individuals with specialized expertise in a

15 particular technical area. See, e.g., Am. Tunaboat Ass’n v. Baldrige, 738 F.2d 1013, 1016-17 (9th Cir.

16 1984) (agency decision under the Marine Mammal Protection Act was not supported by substantial

17 evidence because agency ignored data that was product of “many years’ effort by trained research

18 personnel”).

19 Courts must uphold a reasonable agency action “even if the administrative record contains

20 evidence for and against its decision.” Modesto Irrigation Dist. v. Gutierrez, 619 F.3d 1024, 1036 (9th

21 Cir. 2010) (quotation and citation omitted). “The court’s task is not to make its own judgment,” because

22 “Congress has delegated that responsibility to the [agency].” River Runners, 593 F.3d at 1070. Instead,

23 “[t]he court’s responsibility is narrower: to determine whether the [agency’s action] comports with the

24 requirements of the APA . . . .” Id. The Ninth Circuit has held that “[t]he [agency’s] action . . . need

25 2 The APA does not allow a reviewing court to overturn an agency decision because it disagrees with the

3 decision or with the agency’s conclusions about environmental impacts. Id. This is especially true in

4 the context of management of Forest Service lands, for Congress has consistently acknowledged that the

5 agency must balance competing demands in managing National Forests. See United States v. New

6 Mexico, 438 U.S. 696, 716 n. 23 (1978).

7 III. FACTUAL BACKGROUND

8 Stanislaus National Forest is located between Lake Tahoe and Yosemite National Park on

9 approximately 900,000 acres on the western slope of the Sierra Nevada Mountains. Among the active

10 grazing allotments within Stanislaus National Forest are the Bell Meadow allotment, Eagle Meadow

11 allotment, and Herring Creek allotment (together, “BEH allotments”), which collectively total

12 approximately 51,200 acres. AR 8248.4 Elevation on the BEH allotments varies from approximately

13 6,600 feet to over 9,000 feet. Id. Approximately 75% of the land is forested, with the remaining portion

14 consisting of meadows, open ridge, rock outcroppings, and slope areas. Id. Permitted livestock forage

15 primarily in meadows and riparian areas but also some forested areas. Id. The BEH allotments also

16 have high recreational use year-round, including scenic viewing, backpacking, hiking, fishing, camping,

17 horseback riding, and skiing. Id.

18 The grazing permit for the Bell Meadow allotment was issued on November 15, 2016, AR 13-19,

19 and modified on September 28, 2017, AR 1-4. The permit for the Eagle Meadow allotment was issued

20 on March 28, 2012, AR 27- 33, and modified on September 19, 2017, AR 5-8. The permit for the

21 Herring Creek allotment was issued on June 20, 2016, AR 20-26, and modified on August 24, 2017, AR

22 9- 12. The most recent AOIs were issued in May and June of 2018. AR 88-95 (Bell), AR 81- 87

23 24

25 4 All references t o the administrative record contain the prefix “AR.” 2 In February 2016, the Forest Service released an EIS describing a proposal by the Stanislaus

3 National Forest that would reauthorize livestock grazing on the BEH allotments on the Summit Ranger

4 District. AR 8188. A Draft Record of Decision, issued at the same time, proposed to adopt the

5 alternative described in the EIS that would have reauthorized grazing while implementing an adaptive 6 management strategy5 to ensure compliance with the governing Forest Plan. AR 8150. Approximately

7 six and a half months later, after receiving public comment, the Forest Supervisor withdrew the Draft

8 Record of Decision on August 29, 2016, “[i]n order to allow more interactions with stakeholders on the

9 issues.” AR 7780. That letter also withdrew the proposed plan to reauthorize grazing subject to an

10 adaptive management policy. Id.

11 Livestock grazing on the lands on the BEH allotments stretches back to the 1850s, AR 8248-49,

12 and “[t]he BEH meadows have had a history of grazing and overgrazing going back to the 1890s.” AR

13 8344. See also AR 362 (1965 Management Plan for Bell Meadow, stating “Many of the present and

14 past conflicts between grazing and other renewable resources are in evidence on this allotment,”

15 including “the lack of primary plants in meadow types[] and the serious erosion along Bell Creek. All

16 of these conditions are chronologically successive and can be attributed to overgrazing.”); AR 8829

17 (“Range inspection notes from 1953 indicated active erosion along creeks in Hammill Canyon and that

18 Castle, Coyote, and Wire Corral have been ‘fed to the bone. Absolutely nothing remains.’”).

19 IV. STANDARD OF DECISION

20 Under the APA, the district court’s review of an agency’s decision is usually limited to the

21 administrative record. 5 U.S.C. § 706; see also County of Los Angeles v. Shalala, 192 F.3d 1005, 1011

22 (D.C. Cir. 1999) (when reviewing final agency action, the district court is not managing a “garden

23

24 5 The Draft Record of Decision explains that “[a]daptive management uses monitoring to determine if actions prescribed were followed, and adjusts management if changes are needed. An adaptive strategy is particularly suited for dealing with 25 problems involvi ng high levels of uncertainty, limited knowledge, and unpredictability. This approach lends itself to learning 2 material fact” standard for summary judgment normally does not apply in an APA case. San Joaquin

3 River Group Auth. v. Nat’l Marine Fisheries Serv., 819 F. Supp. 2d 1077, 1083-84 (E.D. Cal. 2011).

4 When reviewing an administrative decision under the APA, there are normally no “disputed facts that

5 the district court must resolve.” Occidental Eng’g Co. v. I.N.S., 753 F.2d 766, 769 (9th Cir. 1985).

6 Instead, “the function of the district court is to determine whether or not as a matter of law the evidence

7 in the administrative record permitted the agency to make the decision it did.” Id.; see also City & Cnty.

8 of San Francisco v. United States, 130 F.3d 873, 877 (9th Cir. 1997). “[S]ummary judgment is an

9 appropriate mechanism for deciding the legal question of whether the agency could reasonably have

10 found the facts as it did.” Occidental, 753 F.2d at 770.

11 V. ANALYSIS

12 A. First Claim for Relief: Clean Water Act

13 Plaintiffs’ APA claims relating to the CWA are in three parts. First, Plaintiffs contend that the

14 Forest Service’s authorization of grazing on the BEH allotments has caused violations of water-quality

15 standards for fecal coliform bacteria, in violation of the Central Valley Regional Water Quality Control

16 Board Basin Plan. Second, they allege that the Forest Service has failed to file a report with the

17 appropriate state board concerning the discharges of waste from livestock grazing on the BEH

18 allotments. Finally, Plaintiffs allege that the Forest Service has initiated new discharges of waste or

19 made material changes in the discharges on the BEH allotments before filing a report and before

20 obtaining a permit or waiver from the permit requirement.

21 Defendants’ arguments do not necessarily track the claims in a neat fashion but are generally

22 summarized as follows. Federal Defendants argue that the Forest Service had no duty to seek a permit

23 prior to authorizing grazing on the BEH allotments and that the decision to authorize grazing on the

24 allotments was neither arbitrary and capricious nor in violation of law. Defendant-Intervenors join in

25 Federal Defendants’ arguments that the Forest Service has complied with the CWA. Defendant- 2 dismissed because “Plaintiffs cannot enforce a grazing regulatory framework that is in the process of

3 being developed” or stayed “pending completion of that process by the Regional Board.” ECF No. 86-1

4 at 5.

5 1. Background Relevant to CWA Claims

6 Evaluating the CWA claims requires an understanding of the complex regulatory framework and

7 factual information from the record. The Porter-Cologne Act is “the principal law governing water

8 quality control in California.” AR 5664. Passed in 1969, “[i]ts goal is ‘to attain the highest water

9 quality which is reasonable, considering all demands being made and to be made on those waters and the

10 total values involved, beneficial and detrimental, economic and social, tangible and intangible.’” City of

11 Burbank v. State Water Res. Control Bd., 35 Cal. 4th 613, 619 (2005) (quoting California Water Code

12 (“CWC”) § 13000). Porter-Cologne “established the State Water [Board], along with nine regional

13 water quality control boards [], and gave those agencies ‘primary responsibility for the coordination and

14 control of water quality.’” Dep’t of Fin. v. Comm’n on State Mandates, 1 Cal. 5th 749, 755 (2016)

15 (quoting CWC § 13001), as modified on denial of reh’g (Nov. 16, 2016). “The [State Water Board]

16 adopts State policy for water quality control and statewide water quality control plans in addition to

17 regulations that are binding on the [regional water quality control boards, which] each govern one of the

18 nine hydrologic regions into which California is divided, adopting regional water quality control plans

19 (basin plans) for their respective regions.” AR 5665 (citing CWC §§ 13200, 13201).

20 Porter-Cologne requires the preparation and adoption of basin plans and requires that those plans

21 “consist[] of a designation or establishment of the waters within a specified area,” the beneficial uses to

22 be protected, the water-quality objectives, and an implementation program to achieve the water quality

23 objectives. CWC § 13050(j). Regional Water Boards adopt and amend basin plans “under a structured

24 process involving full public participation and state environmental review,” and the plans are not

25 effective until they receive State Water Board approval. AR 5087. 2 affect water quality.” AR 5665 (citing CWC §§ 3000, 13050(i), 13140, 13142, 13241). “The Porter-

3 Cologne Act provides that ‘All discharges of waste into the waters of the State are privileges, not

4 rights.’” Id. (quoting CWC § 13263). Both point and non-point source discharges are subject to

5 regulation under Porter-Cologne. Id.

6 The Central Valley Water Board (“Regional Board”) has jurisdiction over the streams that flow

7 through the BEH allotments. AR 8361 (EIS statement that the document used the Central Valley

8 Regional Water Board’s Basin Plan as a regulatory benchmark to assess the effect of the proposed action

9 and its alternatives on water quality). The Regional Board approved the governing Basin Plan in 1998.

10 AR 5076. That Basin Plan states that beneficial uses “are critical to water quality management in

11 California” and are “primary goals of water quality planning.” AR 5092. The Basin Plan includes

12 Water Contact Recreation (termed “REC-1”), which includes “[u]ses of water for recreational activities

13 involving body contact with water, where ingestion of water is reasonably possible,” such as swimming,

14 wading, and fishing, among other activities. Id. The water quality standards for bacteria for REC-1 in

15 the Basin Plan state that “[i]n waters designated for contact recreation (REC-1), the fecal coliform

16 concentration based on a minimum of not less than five samples for any 30-day period shall not exceed a

17 geometric mean of 200/100 ml, nor shall more than ten percent of the total number of samples taken

18 during any 30-day period exceed 400/100 ml.” AR 5104.

19 Section 303(d) of the CWA requires that each state compile a list of waters (known as a § 303(d)

20 list) within the state that do not meet applicable water quality standards. 33 U.S.C. § 1313(d). See also

21 AR 7388 (State Water Resources Control Board document explaining development of Section 303(d)

22 list (citing 40 C.F.R. § 130.7(c), (d))). Section 305(b) of the CWA requires that each state prepare a

23 biennial water quality assessment report concerning all navigable waters in that state. 33 U.S.C.

24 § 1315(b)(1). “Along with the § 303(d) list, the state must submit, inter alia: (1) ‘[a] description of the

25 methodology used to develop the list’; (2) ‘[a] description of the data and information used to identify 2 information’ for certain categories of water; and (4) ‘[a]ny other reasonable information requested by the

3 Regional Administrator.’” Thomas v. Jackson, 581 F.3d 658, 662 (8th Cir. 2009) (quoting 40 C.F.R. §

4 130.7(b)(6)). The Section 303(d) list applies to waters impaired by both point sources and nonpoint

5 sources of pollution. AR 6659.

6 The Regional Water Boards propose recommendations for water bodies to be included on the

7 Section 303(d) list to the State Water Board, which in turn may receive public comments concerning the

8 recommendations and make changes to the recommendations list. AR 7413. The State Water Board

9 then evaluates the data to determine whether standards have been exceeded. AR 7411. States

10 developing their Section 303(d) list are “required to assemble and evaluate all existing and readily

11 available water quality-related data and information, including, at a minimum: (1) waters identified as

12 partially meeting or not meeting designated uses or as threatened in the state’s most recent CWA Section

13 305(b) report; (2) waters for which dilution calculations or predictive modeling indicate nonattainment

14 of applicable standards; (3) waters for which water quality problems have been reported by

15 governmental agencies, members of the public, or academic institutions; and (4) waters identified as

16 impaired or threatened in any CWA Section 319 nonpoint source assessment submitted to the EPA.”

17 AR 6659-60 (citing 40 C.F.R. § 130.7(b)(5)). States must regulate water body segments placed on the 18 Section 303(d) list through development of total maximum daily loads (“TMDLs”)6 or another

19 regulatory step “reasonably expected to result in the attainment of water quality standards within a

20 specified timeframe.” AR 7095.

21 The California State and Regional Water Boards prepare a single integrated report that meets the

22 reporting requirements set forth in Sections 303(d) and 305(b) of the CWA. AR 7095. In 2004, the

23

24 6 “A TMDL is the sum of the individual waste load allocations for point sources, load allocations for nonpoint sources, and 25 natural backgrou nd.” AR 7097 (citing 40 C.F.R. § 130.2(j)). 2 Section 303(d) list for California, including “requirements for data quality, data quantity, and

3 administration of the listing process.” Id. This policy includes factors to be considered for listing and

4 delisting decisions with respect to various pollutants, including bacterial water quality standards, among

5 others. AR 7097. California’s listing policy also requires that assessments of water quality and listing

6 decisions be documented in water body “fact sheets” that include “lines of data” summarizing the

7 available data for a water body segment relative to the applicable standard. AR 7097-98. After

8 reviewing the fact sheets, staff members make listing recommendations.

9 Nonpoint source discharges in California are normally managed through permits or waivers of

10 the permit requirement granted by regional water boards. See AR 5665. The State Water Board’s Water

11 Quality Enforcement Policy has a “progressive enforcement” format that consists of an “escalating

12 series of actions that allows for the efficient and effective use of enforcement resources to: (1) assist

13 cooperative dischargers in achieving compliance; (2) compel compliance for repeat violations and

14 recalcitrant violators; and (3) provide a disincentive for noncompliance.” AR 5063. Progressive

15 enforcement is not a universal starting point, including in situations involving “violations by dischargers

16 with a history of noncompliance,” but it is a standard starting point before “escalat[ing] to increasingly

17 more formal, forceful, and serious actions until compliance is achieved.” Id. The operative Basin Plan

18 acknowledges that “the Regional Water Board recognizes that immediate compliance with water quality

19 objectives adopted by the Regional Water Board or the State Water Board, or with water quality criteria

20 adopted by the USEPA, may not be feasible in all circumstances.” AR 5103. Moreover, in

21 circumstances where the Regional Board “determines it is infeasible for a discharger to comply

22 immediately with such objectives or criteria, compliance shall be achieved in the shortest practicable

23 period of time (determined by the Regional Water Board), not to exceed ten years after the adoption of

24 applicable objectives or criteria.” Id.

25 As discussed more fully below, in 1981, the State Board and the Forest Service signed a 2 discharge report or procure a permit. AR 5913. The MAA also obligates the Forest Service to

3 incorporate best management practices (“BMPs”) for protection of water quality. AR 5840.

4 The state legislature amended Porter-Cologne in 1999 to “require the [State Board] to enforce the

5 state’s NPS pollution control program.” AR 5659. The State Board responded in 2004 through its

6 adoption of the Policy for Implementation and Enforcement of the Nonpoint Source Control Program.

7 AR 5661. That policy provides that “all current and proposed NPS discharges must be regulated under

8 [waste discharge requirements], waivers of [waste discharge requirements], or a basin plan prohibition,

9 or some combination of these administrative tools.” AR 5665. In 2009, the State Board determined that

10 the water quality management plan under the MAA should be revised and initiated a process to update

11 the program. AR 3206, 3268, 3478. In 2009, the State Board adopted a resolution directing State Board

12 staff to propose a statewide approach to address activities on National Forest system lands, including

13 grazing. AR 3268. That effort produced a draft statewide waiver that would have waived waste

14 discharge requirements for nonpoint source discharges for certain activities on National Forest system

15 lands in California. AR 3478. That draft statewide waiver was later withdrawn. In 2015, the State

16 Board discontinued the effort at a statewide level and adopted a resolution directing the regional boards

17 to develop their own systems, taking regional differences into account. AR 2930. Since 2016, the

18 Central Valley Regional Board has been working with the Forest Service to develop a discharge permit

19 that would cover forest-management activities in the region, including grazing. Id. The Regional Board

20 anticipates that a permit could be adopted this year or in 2020. The proposed permits would require

21 only the Forest Service and United States Bureau of Land Management to apply for permit coverage, not

22 individuals or grazing permit holders. AR 3406-08.

23 2. First CWA Claim: Violation of REC-1 Standard

24 Plaintiffs’ first CWA claim is that the Forest Service has authorized grazing on the BEH

25 allotments that has resulted in violations of the Basing Plan’s REC-1 standard, an action that was 2 a. CSERC Testing

3 In 2009, CSERC staff prepared a quality assurance project plan (“QAPP”) to establish

4 methodologies to sample, gather, and report on water quality in streams in Stanislaus National Forest.

5 AR 6501-43. That QAPP addressed protocols such as training and lab certification, methodology for

6 collecting water samples, chain of custody for the samples, equipment calibration, and analytical

7 methods. AR 6504-05. A senior environmental scientist from the State Water Board served as a

8 technical advisor to CSERC in the preparation of the QAPP. AR 6504, 7193.7 The State Water Board

9 stated that data collected under a QAPP “pursuant to the requirements of 40 CFR 31.45 are acceptable

10 for use in developing the [CWA] section 303(d) list.” AR 7407.8

11 Between May and August of 2009, CSERC tested streams on each of the BEH allotments, one

12 stream on another allotment within Stanislaus National Forest that authorizes cattle grazing, and a

13 control site (Bourland Creek) where no cattle grazing is authorized. AR 6297-99. Following the QAPP,

14 CSERC took multiple samples within the streams it tested and conducted testing both before and after

15

16 7 The Regional Board stated in response to comments to the Final 2014 Integrated Report that the scientist “is a Senior

17 Environmental Scientist and Coordinator for the Clean Water Team (CWT).” It further explained that:

The CWT is the citizen monitoring program of the State Water Resources Control Board and is part of the 18 Surface Water Ambient Monitoring Program (SWAMP). Mr. Burres’s role as Coordinator is to work statewide to provide technical assistance and guidance documents, training, and QA/QC support to citizen 19 monitoring programs, non-profit groups, Farm Bureaus, Resource Conservation Districts, Coordinated Resource Management and Planning groups, government agencies including the U.S. Forest Service, 20 Tribes, and colleges. Mr. Burres provided telephone consultation to the Central Sierra Environmental Resource Center regarding how to design a monitoring plan and QAPP, and how to select and follow the SWAMP methods and procedures for data collection and analysis, to enable SWAMP-level quality and 21 comparability so that resulting data can be used in Water Board assessments. For more information about the Clean Water Team, please refer to this Water Board website:

22 http://www.waterboards.ca.gov/water_issues/programs/swamp/cwt_volunteer.shtml.

AR 7193.

23

8 40 C.F.R. § 31.45 provides: “If the grantee’s project involves environmentally related measurements or data generation, the 24 grantee shall develop and implement quality assurance practices consisting of policies, procedures, specifications, standards, and documentation sufficient to produce data of quality adequate to meet project objectives and to minimize loss of data due 25 to out-of-control conditions or malfunctions.” 2 California, a state-certified analytical laboratory. AR 6299. AquaLab tested the samples for E. coli, 3 total coliform, and fecal coliform bacteria.9 Id. CSERC’s testing reported 17 violations of REC-1

4 bacteria standards in streams on Lower Round Meadow in the Bell Meadow Allotment, AR 6307-12; 5 seven violations of REC-1 bacteria standards in streams on Barn Meadow10 in the Eagle Meadow

6 Allotment, AR 6301-02; and 11 violations of REC-1 bacteria standards in streams on Bull Run Meadow

7 and Upper Fiddler’s Green Meadow in the Herring Creek Allotment, AR 6304-07. CSERC documented

8 no REC-1 violations in samples taken from the headwaters of Bourland Creek below Bourland Meadow,

9 a control site with no livestock grazing. “In comparison to the significant increase in fecal coliform

10 colonies quantified at the streams with grazing once livestock were present,” the fecal coliform

11 concentration from Bourland Meadow “remained consistently low and within standard limits throughout

12 the same time period that the grazed samples were being collected.” AR 6320.

13 Between May and August of 2010, CSERC tested streams in non-BEH allotments within

14 Stanislaus National Forest, finding 100 REC-1 violations on eight allotments sampled and zero REC-1

15 violations on the two control sites where no grazing was authorized. AR 6232. Like the 2009 report,

16 this study found that fecal coliform bacteria in surface waters were below the REC-1 standard before

17 grazing but increased substantially after grazing began “and in places exceeded state standards” during

18 the summer grazing period. AR 6201.

19 b. CSERC Water Data Incorporated into State Water Board and EPA Processes

20 The Regional Board solicited data for use in development of the Section 303(d) lists from mid-

21 January to August 30, 2010. AR 7184. CSERC submitted the data it collected during the summers of

22

23 9 Although fecal coliform and E. coli are “generally not harmful themselves,” they are indicator species that signal that fecal 24 contamination may have occurred. AR 8364.

25 10 “Barn Meadow lies along the headwaters of Niagara Creek, with segments in contact with the stream.” AR 8371. 2 integrated report for 2014-2016 for public comment beginning in September of 2016.11 AR 7086. The

3 draft report proposed inclusion of six creeks that CSERC recommended, including Bell Creek on the 4 Bell Meadow allotment and Niagara Creek on the Eagle Meadow allotment.12 AR 7204, 7158, 7165.

5 On October 20, 2016, the Forest Service submitted comments to the Regional Board concerning the

6 water bodies proposed to be included on the 303(d) listings. AR 7281-85. The Regional Board also

7 received comments from 11 other commenters. AR 7183.

8 In December 2016, the Regional Board issued its final integrated report that included responses

9 to the comments received during the comment period concerning the water-quality assessments for 10 creeks on or near grazing allotments on Stanislaus National Forest.13 AR 7183-7207. In response to

11 comments questioning the use of CSERC data, the Regional board stated that the CSERC data complied

12 with “minimum requirements for quality control and assurance, temporal and spatial characteristics, and

13 minimum samples sizes established by the Listing Policy and are therefore appropriate for use in the

14 listing process.” AR 7187. The report also stated that the data “were collected during the critical season

15 for the pollutant and applicable water quality standard (spring/summer for recreational beneficial uses)”

16 and that, according to the QAPP, sampling only took place when the water bodies had adequate flow

17

18 11 The process appears to be a lengthy one. The State Water Board adopted the statewide 2012 report in 2015, and the most 19 recent information in that report for the Central Valley was from the 2010 Central Valley Integrated Report. AR 7099. The 2014-2016 draft integrated report noted that “the timeline for completing TMDLs for water bodies listed for the first time as 20 part of the 2014 Integrated Report is estimated to be no longer than thirteen years, which equates to an estimated completion date of 2027.” AR 7116.

21 12 CSERC also recommended inclusion of Bull Meadow Creek, Jawbone Creek, an unnamed tributary to Jawbone Creek, and Rose Creek in Tuolumne County. After its review of the data, the Regional Board did not recommend including Jawbone

22 Creek. AR 7176, 7204.

13 The responses to the comments are grouped by frequency of appearance in the comment letters. Ten of the comments 23 appeared in multiple comment letters, which included the comments that grazing was not the only source of the bacteria; that the data was collected prior to August, 2010, and should be updated before a listing decision was made; that the assessments 24 incorporated only data from one source, CSERC; that there was potential bias in the collection of samples; and that the samples are not representative of overall water-quality conditions. The final report also responded to comments raised only 25 in the USFS’s co mments. AR 7196-7200. 2 because “the water bodies are small ephemeral creeks with limited swimming areas,” reiterating instead

3 that the beneficial uses of the Tuolumne and Stanislaus Rivers apply to their upstream tributaries that do

4 not have separately listed beneficial uses in the basin plan. AR 7190-91. The Regional Board also

5 rejected the argument that the water-quality standards should be assessed under the EPA’s new 2012

6 standards, making two primary points. First, the Regional Board used the water-quality objective for

7 fecal coliform bacteria under the governing basin plan, and indicated that “[d]etermining the

8 appropriateness of water quality objectives and revising the Basin Plan” was “outside the scope of this 9 listing process.”14 AR 7192. Second, using the EPA’s 2012 standards (which were in draft form at the

10 time the project began) “would likely increase the number of exceedances identified in the lines of

11 evidence” because the standard in the 2012 criteria is “more stringent” than the criteria in the EPA’s

12 1986 criteria. AR 7191. The Regional Board agreed with the comment raised by all but three of the

13 commenters that “grazing should not be considered the sole source of bacteria to these streams since

14 there are wildlife species and other potential sources,” but concluded that “since the available data and

15 information indicate that grazing animals are a likely potential source of indicator bacteria to these

16 streams,” the proposed 303(d) listings for the creeks would continue to identify them as “grazing related 17 source.”15 AR 7185. The final integrated report continued to propose to add to the 303(d) list five of the

18 six streams listed in the draft report, including Bell Creek on the Bell Meadow allotment and Niagara

19 Creek on the Eagle Meadow allotment. AR 7158, 7165; see also AR 8355 (stating that 6.6 miles of Bell

20 Creek is on the Bell Meadow allotment); AR 8366 (listing Niagara Creek on the Eagle Meadow

21

22 14 Although the EPA in 2012 began recommending that E. coli rather than fecal coliform be used “as the best indicator of 23 health risk from water contact in recreational waters,” the Regional Board has not changed the basin plan’s REC-1 standard and continues to use fecal coliform as the standard indicator. AR 8364. 24 15 The Regional Board modified the listings to add on the list of potential sources “natural sources” and “source unknown” 25 even as it continu ed to identify grazing as a potential source as well. Id. 2 The State Water Board solicited public comments on the proposed list beginning in June 2017

3 and held a public hearing on the list on October 3, 2017, approving the list that same day. AR 6662. 4 The approved 303(d) list included both Bell Creek and Niagara Creek.16 AR 6730. On April 6, 2018,

5 EPA approved the State Water Board’s list of impaired waters, “including all water quality limited

6 segments (WQLSs) and associated pollutants identified by the State Water Resources Control Board

7 (State Board) as requiring a total maximum daily load under CWA section 303(d).” AR 6658.

8 c. Recent CSERC Monitoring

9 CSERC has continued to test water quality in streams on Stanislaus National Forest land in the

10 years since 2009-10, including streams on the BEH allotments, using the same methodology that the

11 Regional and State Boards relied on to include Bell Creek and Niagara Creek on the 303(d) list.

12 • Between June and August 2011, CSERC tested streams in Stanislaus National Forest. Because of heavy snowpack, no livestock grazed on the Eagle Meadow or Herring Creek allotments that

13 grazing season. AR 6191. The CSERC report did not calculate the number of violations found on the Lower Round Meadow on the Bell Meadow allotment but noted “a rapid rise” in the fecal

14 coliform and E. coli concentrations after cattle arrived to graze. AR 6192. • Between June and August 2012, CSERC documented 21 bacterial violations17 on the Lower

15 Round Meadow on the Bell Meadow allotment. AR 6174-79.

• Between June and August 2013, CSERC documented 12 bacterial violations on two wilderness 16 sites not on the BEH allotments that had some livestock grazing. AR 6170, 6183-86. • Between June and September 2014, CSERC documented 27 bacterial violations on Bell Creek 17 on the Bell Meadow allotment, zero violations on Niagara Creek on the Eagle Meadow allotment, and zero violations on a major tributary of Cow Creek on the Herring Creek allotment. 18 AR 6155, 6156, 6161-6166. • Between May and September 2015, CSERC documented 49 bacterial violations on Bell Creek 19 on the Bell Meadow allotment, the only site tested on the BEH allotments.18 AR 6118-18, 6141- 52. 20 • Between June and October 2016, CSERC documented 20 bacterial violations on Bell Creek on the Bell Meadow allotment, seven bacterial violations on Cow Creek on the Herring Creek 21

22 16 The State Water Board’s supporting information for each creek listed the expected TMDL completion date as 2027. AR 6731, 6739. 23

17 Unless otherwise noted, “bacterial violations” here refers only to fecal coliform measurements, though the CSERC reports 24 also frequently sample for E. coli.

25 18 CSERC also d ocumented 85 violations on four streams in other allotments not at issue in this case where livestock grazing 2 6 A1 R0 6 6. 1 0C 4S . E RC documented no bacterial violations at the tributary of Bell Creek that it tested. • Between May and November 2017, CSERC documented two bacterial violations on Bell Creek

3 on the Bell Meadow allotment, zero bacterial violations on Cow Creek on the Herring Creek allotment, five bacterial violations on Herring Creek in Hammill Canyon Meadow on the Herring

4 Creek allotment, and two violations on Niagara Creek on the Eagle Meadow allotment. AR 6067-68, 6070-72, 6080-81.

5 Plaintiffs summarize the reports to mean that between 2009 and 2017, CSERC data demonstrate

6 136 violations of the REC-1 standard for fecal coliform on the Bell Creek allotment, 12 on the Eagle

7 Meadow allotment, and 23 on the Herring Creek allotment. ECF No. 87 at 11.

8 Plaintiffs argue that the same testing that CSERC conducted in 2009-2010 that the Regional

9 Board, State Board, and EPA relied on for purposes of a 303(d) listing has been continued in the years

10 since and have continued to show that the Forest Service is authorizing grazing on its land that results in

11 violations of the REC-1 standard in the Basin Plan, in violation of the APA. See id. at 10-11.

12 d. Programmatic Challenge and Attenuated Causal Chain

13 Federal Defendants first contest whether Plaintiffs can bring this sort of claim at all, arguing that

14 the general nature of the challenge is inadequately tied to specific agency action and that the alleged

15 violation occurs too far down the causal chain for Federal Defendants to be responsible.

16 Federal Defendants argue that Plaintiffs have not identified a specific “final agency action,” as

17 they must when bringing a challenge under the APA, and that by lumping the grazing permits and AOIs

18 together, this suit is an impermissible “programmatic challenge,” rather than a specific agency action

19 that has “an actual or immediately threatened effect.” ECF No. 85-1 at 9 (quoting Lujan v. Nat’l

20 Wildlife Fed’n, 497 U.S. 871, 892-94 (1990)). The Court agrees that the TAC does seek relief in broad

21 terms, but the violations identified in the various CSERC reports that form the basis of the CWA claim

22 focus mostly on Bell Creek on the Bell Meadow allotment. The other water violations likewise took

23 place on a handful of allotments, making the challenge more discrete than it appears at first blush. In

24 any case, the three dozen allotments on Stanislaus National Forest land is not so broad and unwieldy that

25 2 Federal Defendants also take another gentle pass at the argument that this action is barred by

3 sovereign immunity, an argument the Court rejected in its order on Federal Defendants’ motion to

4 dismiss. ECF No. 55 at 22. Federal Defendants now argue more forcefully that even if the CWA claim

5 is not barred by sovereign immunity, the causality alleged in the TAC is too attenuated to survive.

6 Federal Defendants maintain that the Forest Service had no duty to seek a permit because an agency

7 authorizing legal activity does not incur APA liability for that action when the alleged violation is at the

8 long end of a causal chain. They cite Protect Our Communities Foundation v. Jewell, 825 F.3d 571 (9th

9 Cir. 2016) (“POCF”), in support of their argument that the causal chain between the Forest Service

10 action and the violation alleged in the TAC is too attenuated to be actionable under the APA. The

11 plaintiffs in POCF challenged the Bureau of Land Management’s (“BLM”) decision to grant a right-of-

12 way on federal land that would permit a company to build and operate a wind energy facility on 12,360

13 acres. BLM issued the right-of-way only after adopting a final EIS modifying the plan to reduce the risk

14 of birds colliding with wind turbines blades and further provided that the grant was conditioned on the

15 adoption of certain “mitigation measures and monitoring programs.” Id. at 577. Relatedly, certain of

16 the bird species at risk were protected by the Migratory Bird Treaty Act (“MBTA”), and the Bald and

17 Golden Eagle Protection Act (“BGEPA”). The U.S. Fish and Wildlife Service (“FWS”) —the agency

18 charged with implementing relevant aspects of the MBTA and BGEPA—advised BLM that the

19 collection of mitigation measures BLM adopted were not the equivalent of a “take permit” that would

20 have authorized harming birds under the MBTA and/or BGEPA. Yet FWS acknowledged that the

21 mitigation measures “could serve as the basis for a future permit application with the FWS.” Id. at 578.

22 A group of plaintiffs brought an APA challenge to the right-of-way grant as unlawful under NEPA,

23 MBTA, and BGEPA. Plaintiffs put forward two theories of liability. First, they argued that BLM,

24 “acting in its regulatory capacity,” was directly liable for unlawful “take” of birds under the Acts

25 without a permit from Fish and WS. Id. at 585. Second, plaintiffs argued that the authorization was 2 “securing the appropriate permits from the FWS.” Id.

3 The Ninth Circuit affirmed the grant of summary judgment to the federal defendants on both

4 counts. The court first held that the MBTA “does not contemplate attenuated secondary liability on

5 agencies like the BLM that act in a purely regulatory capacity, and whose regulatory acts do not directly

6 or proximately cause the ‘take’ of migratory birds” within the definition set forth in that statute. Id. at 7 585.19 More directly relevant to this case, the court held that the APA claim failed because “the BLM’s

8 regulatory role in this case is too far removed from the ultimate legal violation to be independently

9 unlawful under the APA.” Id. at 586. Federal Defendants argue that the same result should issue here,

10 where there is the risk of “unbounded agency vicarious liability” if the Forest Service is found liable. Id.

11 The agency action in POCF was “far too removed from the ultimate legal violation to be

12 independently unlawful under the APA.” Id. The court held the BLM merely granted a permit “to

13 engage in otherwise lawful activities that would incidentally lead to migratory-bird deaths” and that the

14 challengers’ argument “hinge[d] on the assumption” that the third-party permittee will operate wind

15 turbines in an unlawful manner. Id. With BLM having done nothing more than authorize the

16 development of a wind-energy facility, and “[w]ithout further indication of its involvement in the

17 putative violation,” the court declined to “hold the BLM complicit in future unlawful activity, separately

18 committed by a grantee, through a mere failure to intervene at the permitting stage.” Id. at 587.

19 Moreover, the BLM approval was contingent on the permittee’s compliance with the law, including the

20 securing of necessary permits, and the terms of the right-of-way permitted the BLM to withdraw

21 approval if it determined that the permittee was not complying with those provisions. Id.

22

23 19 The court distinguished the case of the National Marine Fisheries Service’s decision to apply for a permit from the FWS to cover incidental take of migratory seabirds by a Hawaii fishery. The court surmised that the agency there was functioning in 24 a “managerial capacity over the activities of the fishery,” meaning that it “occup[ied] a more directly supervisorial position over a regulated third party than that of a typical agency” or, alternatively, was playing it safe by deciding to shield itself 25 from potential lia bility by applying for a permit it likely was not compelled to secure. Id. at 586. If the former, the closeness 2 closer and more active. As explained above, the Forest Service issues permits and AOIs that determine

3 the type of species, the number, the areas of permissible grazing, and dates during which grazing may

4 occur. As discussed more fully below in the context of the NFMA claims, the Forest Service engages in

5 monitoring of the allotments to ensure compliance with the operative Forest Plan. It has adopted an

6 adaptive management framework to monitor compliance with the plan and to modify the grazing on the

7 BEH allotments to ensure movement toward achievement of the Forest Plan’s objectives. The Forest

8 Service is empowered to and does modify the permits and AOIs that exerts a level of control over the

9 permittees that was absent in POFC. The Forest Service has, for example, recently modified grazing

10 permits to include maps with special aquatic features on the permittees’ allotments that should be

11 protected, AR 3, 7; has modified AOIs to include mitigation measures like installing an electric fence,

12 installing a water trough on a specific meadow, minimizing use of another specific meadow, and herding

13 livestock away from yet another meadow, AR 121, 133; and has installed an electric fence to prevent

14 overgrazing of young aspen, AR 607-08. This is a granular level of control that was absent in POFC but

15 is substantially similar to the “directly supervisorial position” an agency could occupy that would

16 distinguish it from the attenuated role at issue in POFC. 825 F.3d at 586.

17 The POCF court distinguished a case involving an agency’s compliance with the Endangered

18 Species Act because there was no statutory duty in POCF that created any “affirmative duty on the part

19 of the BLM to guarantee a grantee’s compliance.” Id. at 587. In contrast, the alleged CWA violations

20 here do involve just that sort of agency duty that was absent in POCF. CWA § 313 requires that federal

21 agencies like the Forest Service comply with state water standards to the same extent as nongovernment

22 agencies. Under the 1981 MAA between the State Board and the Forest Service, the Forest Service

23 agreed to be the designated water-quality management agency for national forest lands in California.

24 AR 6029. The MAA obligated it to incorporate BMPs for protection of water quality and to monitor the

25 program’s effectiveness. AR 5840-41. More recently, the Regional Board is in the process of crafting a 2 discharge permit that would cover forest-management activities in the region, including grazing, AR

3 2930, and would require only the Forest Service and United States Bureau of Land Management to

4 apply, not individuals or grazing permit holders. AR 3406-08. The level of control that the Forest

5 Service has over grazing activities and its role as the more directly regulated party with respect to state

6 water regulators brings this case far outside the attenuated speculative causal chain at issue in POFC.

7 The Court agrees that an agency acting in a mere regulatory role “far removed from the ultimate

8 legal violation” can be too attenuated to violate the APA. But, the violations alleged here are closely

9 tied to agency action and the agency itself is the entity whose conduct is regulated.

10 e. Forest Service’s Close Contact with State Regulators

11 Federal Defendants next argue that Forest Service permitting of grazing on the BEH allotments

12 has not violated the APA. Focusing on the forgiving standard of review, Federal Defendants argue that

13 the Forest Service’s action taken in close engagement with state water agencies prevents a finding that

14 the agency acted arbitrarily, capriciously, or in violation of law, even where there were exceedances of

15 bacteria levels.

16 In Center for Native Ecosystems v. Cables, 509 F.3d 1310 (10th Cir. 2007), like here, the

17 plaintiffs challenged the issuance of AOIs as arbitrary and capricious under the APA and in violation of

18 CWA § 313(a), 33 U.S.C. § 1323(a), because nonpoint-source pollution had resulted in levels of fecal-

19 coliform bacteria in violation of Wyoming regulations. The bacterial violations led the state of

20 Wyoming to list the waters as impaired under the CWA. The court concluded that under Wyoming’s

21 water-quality standards, nonpoint-source pollution was “not under ‘control’” and accordingly subject to

22 state enforcement action only if BMPs had not been implemented. Id. at 1332. The court held that “so

23 long as BMPs have been implemented, the state agency has no authority to take enforcement action, and

24 the Forest Service cannot be said to have failed to comply with state requirements ‘in the same manner,

25 and to the same extent as any nongovernmental entity.’” Id. at 1333 (quoting 33 U.S.C. § 1323(a)). 2 understanding with local conservation districts, the court determined that the Forest Service

3 demonstrated no “clear error of judgment” in issuing the AOIs. Id.

4 In Center for Biological Diversity v. Wagner, No. CIV. 08-302-CL, 2009 WL 2176049, at *16

5 (D. Or. June 29, 2009), report and recommendation adopted, No. CIV. 08-302-CL, 2009 WL 2208023

6 (D. Or. July 22, 2009), the court held that in an APA challenge brought pursuant to CWA § 313, Oregon

7 law had a safe-harbor provision for designated management agencies that had implemented best

8 management practices, such that even evidence of elevated E. coli levels did not mean that the Forest

9 Service had acted arbitrarily or capriciously. Id. at *18. The Wagner plaintiffs pointed to Northwest

10 Indian Cemetery Protective Association v. Peterson, 795 F.2d 688, 697 (9th Cir. 1986), rev’d on other

11 grounds sub nom. Lyng v. Northwest Indian Cemetery Protective Association, 485 U.S. 439 (1988), a

12 Ninth Circuit case involving the Forest Service’s alleged failure to comply with water-quality standards

13 in a California regional water board’s basin plan. In Peterson, the Ninth Circuit rejected the argument

14 that implementation of BMPs gave the Forest Service a safe harbor, because the BMPs “are merely a

15 means to achieve the appropriate state Plan water quality standards,” finding “no indication in the Plan

16 or in the agreements between the Forest Service and the Water Quality Control Board that the BMPs

17 were to be considered standards in and of themselves,” and concluding “[a]dherence to the BMPs does

18 not automatically ensure that the applicable state standards are being met.” Id. The Wagner court

19 distinguished Northwest Indian Cemetery Protective Association, noting that unlike the basin plan at

20 issue there, the Oregon regulations in Wagner did contain a safe-harbor provision providing that a

21 management agency’s adoption of BMPs satisfied state water-quality standards. The court granted

22 summary judgment to the Forest Service because it had implemented those BMPs and accordingly did

23 not violate the APA.

24 Oregon later removed the provision permitting BMPs to serve as compliance with state water-

25 quality standards, and the Forest Service again faced a CWA § 313 challenge for failing to comply with 2 Forest Service and state regulators had entered into a memorandum of understanding, and the Forest

3 Service had developed a water quality restoration plan that state regulators found adequately addressed

4 its obligations. Id. at 1169-70. Under these circumstances, even without the BMP safe-harbor, the court

5 found that because the Forest Service “designed and implemented multiple measures to ensure

6 achievement of temperature standards” and state regulators had “approved of those efforts and certified

7 their compliance with state regulations,” the Forest Service did not act in an arbitrary and capricious

8 manner. Id. at 1171. The court concluded that “[e]xceedances are expected and do not render the Forest

9 Service’s efforts invalid. The Forest Service rationally issued the challenged AOIs on this basis.” Id.

10 The lesson of these cases is even when an agency might be shown to otherwise technically

11 violate a state water quality standard, compliance can be achieved in a number of indirect ways,

12 including through the implementation of BMPs serving as an acknowledged alternative to direct 13 compliance;20 or through the existence of other mechanisms, including a combination of BMPs,

14 memoranda of understanding with the state regulators, and other efforts designed to achieve compliance

15 with water quality standards

16 Two elements of Oregon Wild and Cables are instructive. In Oregon Wild, the state regulators

17 acknowledged that “violations may occur while the Forest Service works to achieve long-term goals,”

18 but certified that the Forest Service’s efforts as outlined in a water-quality plan and in memoranda of

19 understanding would serve to satisfy their obligations to comply with state water-quality standards.

20 Oregon Wild, 193 F. Supp. 3d at 1170. Similarly, in Cables, the Tenth Circuit rejected the claim that

21 evidence of continued bacterial violations even after implementation of BMPs violated the law, because

22 “that is not the standard dictated by state regulations and the CWA.” Cables, 509 F.3d at 1333

23 24

25 20 No Defendant here has identified a best management practices provision that excuses violations of the applicable water 2 state agency would work with polluters to identify modifications to the best management practices. Id.

3 Because the state regulators lacked authority to take enforcement action so long as the BMPs were being

4 implemented, the Forest Service was complying with state water law to the same extent as a

5 nongovernmental actor and was accordingly not acting arbitrarily or capriciously. Id. This was so even

6 though the alleged water-quality violations led to a listing of the water bodies as impaired under the

7 CWA. See id. at 1319.

8 Unlike in Oregon Wild, state regulators here have not “certified” that the Forest Service’s efforts

9 comply with state regulations, and unlike in Cables, the state agency here does have authority to take

10 enforcement action despite implementation of best management practices. Yet, Federal Defendants

11 emphasize that the regulatory process stretching back to the 1981 signing of the MAA has kept the

12 Forest Service in close engagement with the State and Regional Boards concerning water quality on

13 Forest Service land. See ECF No. 85-1 at 14-20. The MAA obligated the Forest Service to implement

14 BMPs for water quality and to monitor the effectiveness of those efforts and exempted the Forest

15 Service from filing notices of discharge and applying for a permit or waiver from the permit

16 requirement. AR 6028-34. Since the State Board determined in 2009 that the water quality

17 management plan under the MAA should be revised, there has been a lengthy effort to work with

18 stakeholders to update the regulatory scheme for nonpoint source activities on National Forest land. See

19 AR 3206, 3268, 3478 The close relationship between the Forest Service and the State and Regional

20 Boards weighs against a finding that the Forest Service has acted in violation of the APA, especially in

21 light of the State Board’s progressive enforcement structure and the Basin Plan’s acknowledgement that

22 “that immediate compliance . . . may not be feasible in all circumstances.” AR 5103. Viewing the

23 circumstances as a whole, where the MAA has required the implementation and monitoring of water

24 quality, state regulators have for the past decade worked to craft an updated regulatory framework for

25 nonpoint source discharges on Forest Service land, and Plaintiffs have not identified any formal or 2 regulations alleged to have been violated, the Forest Service’s action do not appear to have been 3 arbitrary, capricious, or contrary to law, in violation of the APA.21

4 f. Forest Service’s Steps to Address Water Quality In Grazing

5 The conclusion that the Forest Service’s decision to issue grazing permits and AOIs did not

6 violate the APA is bolstered by the actions the Forest Service has taken to address water quality.

7 Federal Defendants argue that the Forest Service has not acted arbitrarily, capriciously, or in violation of

8 law because it rationally addressed water quality in its grazing authorizations, pointing to steps it has

9 taken to modify the permits and AOIs. ECF No. 85-1 at 21-21.

10 First, in May 2017, the Forest Service incorporated provisions relating to water quality into the

11 grazing permits. See AR 3, 7, 11. The language the Forest Service identifies states, in a section labeled

12 “Water Quality” in a grazing permit modification:

13 One of the Forest Goals described in the Forest Plan Direction is to “Maintain or improve water quality and watershed conditions to meet applicable state and federal requirements,”

14 including water quality standards established by the state and regional water boards. Forest Plan Direction p. 5. Meeting riparian S&Gs for forage use and streambank disturbance on

15 the Bell Meadow and Upper Hull allotments will facilitate this goal, as will herding, salting and other management practices that reduce cattle concentration in riparian areas and

16 increase upland distribution. 17 AR 3 ((grazing permit modification dated September 2017). The Forest Service also points to language 18 included in the 2017 and 2018 AOIs that is identical but for the identification of specific allotments 19 identified in the last sentence. AR 78, 85, 92, 100, 108, 115. 20 Second, the Forest Service points to its efforts to improve water-quality infrastructure. The 21 Forest Service purchased three livestock water troughs in 2017 with plans to install them on the Herring 22 Creek allotment in 2018. AR 5782. The purpose of the troughs is to provide water to the livestock

23

24 21 The Regional Board did comment on the EIS and register its disagreement with the Forest Service’s comments concerning the Basin Plan. See AR 9111 (letter dated Mar. 12, 2014, commenting on draft EIS and objecting to statement in draft EIS 25 that high bacteria l concentrations were “discounted” because they occurred in areas where ingestion of water was unlikely, 2 point to language included in the 2018 AOIs for Bell Meadow and Eagle Meadow allotments

3 acknowledging that the EPA approved the State Board’s inclusion of Bell Creek and Niagara Creek on

4 the 303(d) list of impaired waters, repeating the recommendations included in the permit modifications

5 about using herding and salting to reduce cattle distribution in riparian areas. AR 85, 92. The 2018 Bell

6 Meadow AOI also mentions as a priority the installation of a solar pump and watering trough at Middle

7 Bell Meadow to reduce livestock presence near the stream. AR 92; see also AR 5782 (discussing the

8 stream’s water gap creating concentration of livestock near Bell Creek on Middle Bell Meadow and why

9 water trough will help to reduce this concentration).

10 Third, Federal Defendants point to the Forest Service’s partnership with the University of

11 California, Davis (“UC Davis”), which began in 2010. AR 6009. This partnership formed to sample

12 water quality to examine how grazing was affecting water quality on the BEH allotments. AR 8363-64.

13 The UC Davis researchers published their findings in a peer-reviewed article in June 2013, presenting

14 their findings from the 2011 water quality sampling in Stanislaus National Forest. AR 5986. In August

15 2016, UC Davis scientists indicated that the Regional Board had expressed interest in forming a

16 partnership to develop a strategy to protect water quality and compliance with regulatory standards on

17 grazing land. AR 5972-73, 5978. UC Davis also began weekly sampling on Stanislaus National Forest

18 in July of 2016 and planned to continue through September of that year. AR 5972.

19 Finally, Federal Defendants cite the Forest Service’s own monitoring, which began in earnest in

20 2011. AR 525.

21 Together, the Forest Service contends that these actions demonstrate that the Forest Service has

22 considered water quality on the BEH allotments and modified its livestock grazing program in response.

23 ECF No. 85-1 at 22. In light of the close regulatory scheme, the Forest Service’s efforts to monitor

24 bacteria levels and limit the likelihood of cow waste ending up in water bodies near the BEH allotments

25 bolsters the conclusion that the Forest Service has not acted in violation of the APA by authorizing 2 In light of the authorities discussed above and given the ongoing, working regulatory relationship

3 between the Forest Service and the Regional and State Boards and the Forest Service’s efforts to reduce

4 potential water quality violations, issuing the challenged grazing permits and AOIs was not arbitrary,

5 capricious, or contrary to law. Cables, 509 F.3d at 1333 (10th Cir. 2007) (holding that Forest Service

6 efforts to modify livestock grazing to limit fecal-coliform bacteria violations, along with “[t]he Forest

7 Service’s ongoing implementation of BMPs and its entry into a Memorandum of Understanding with

8 local conservation districts reflect a reasoned approach to elevated fecal-coliform levels” that did not

9 violate the APA).

10 3. Second and Third CWA Claims: Failure to File a Discharge Report or To Procure a Permit or Waiver

11 The second component of Plaintiffs’ claim in the TAC that the Forest Service has violated the

12 APA has two related parts. First, Plaintiffs contend that the Forest Service has unlawfully failed to file a

13 report with the Regional Board. Second, Plaintiffs argue that the Forest Service has unlawfully initiated

14 new discharges of waste or made material changes in such discharges on the BEH allotments before

15 filing a report and before obtaining a permit or waiver from the permit requirement.

16 The Porter-Cologne Act provides that any person “discharging waste, or proposing to discharge

17 waste, within any region that could affect the quality of the waters of the state” is required to file a

18 report with the appropriate regional water board. CWC § 13260. “The regional board then ‘shall

19 prescribe requirements as to the nature’ of the discharge, implementing any applicable water quality

20

21

22 22 The Forest Service also contests the quality of the CSERC data. Sone of the arguments, such as the fact that EPA now recommends use of E. coli rather than fecal coliform bacteria for water-quality testing, were specifically rejected by the 23 Regional Board. Others, such as the fact that a 303(d) listing does not mean that livestock grazing was the sole cause of any bacterial violations (especially in warm-weather months, when bacteria sources apart from cattle can and do contribute to 24 bacteria levels), are well-taken and were acknowledged by the Regional Board. The Court need not resolve this dispute, because, as discussed herein, even assuming the validity of the CSERC data, the Forest Service has not acted arbitrarily, 25 capriciously, or i n violation of law. 2 § 13263(a)), as modified on denial of reh’g (Nov. 16, 2016).

3 Plaintiffs argue that under the plain language of Porter-Cologne, the Forest Service is required to

4 file a discharge report and that the Forest Service admitted as much in response to a comment to the EIS.

5 In response to a comment stating that the Forest Service was not complying with the Clean Water Act or

6 the Porter-Cologne Act in its administration of livestock grazing on the BEH allotments, the EIS stated,

7 “The Forest will continue working with the regional water quality control board to obtain a permit or

8 waiver for nonpoint source pollution in order to ensure compliance with all applicable water quality

9 regulations.” AR 8492. Plaintiffs also assert that the Forest Service has violated CWC § 13264, which

10 provides 11 No person shall initiate any new discharge of waste or make any material changes in any discharge, or initiate a discharge to, make any material changes in a discharge to, or 12 construct, an injection well, prior to the filing of the report required by Section 13260 and no person shall take any of these actions after filing the report but before whichever of the 13 following occurs first:

14 (1) The issuance of waste discharge requirements pursuant to Section 13263.

15 (2) The expiration of 140 days after compliance with Section 13260 if the waste to be discharged does not create or threaten to create a condition of pollution or nuisance and 16 any of [a list of enumerated conditions applies.]

17 . . .

18 (3) The issuance of a waiver pursuant to Section 13269.

19 CWC § 13264. Plaintiffs argue that because the Forest Service has not filed a discharge report and has

20 continued to authorize grazing on the BEH allotments, it has violated CWC §§ 13260 and 13264.

21 The Forest Service and Defendant-Intervenors argue that the Forest Service is under no such

22 obligation because it is operating under the 1981 MAA with the State Water Board. In 1979, the State

23 Board designated the Forest Service as the management agency “for all activities on NFS lands effective

24 upon execution of a management agency agreement.” AR 6029. In the 1981 MAA, the Forest Service

25 agreed that it would be designated the water-quality management agency for national forest lands in 2 That Section 313 of the Federal Water Pollution Control Act mandates federal agency compliance with the substantive and procedural requirements of state and local water 3 pollution control law. It is contemplated by this agreement that Forest Service reasonable implementation of those practices and procedures and of this agreement will constitute 4 compliance with Section 13260, subdivision (a) of Section 13263, and subdivision (b) of Section 13264, Water Code. It is further contemplated that these provisions requiring a 5 report of proposed discharge and issuance of waste discharge requirements for nonpoint source discharges will be waived by the Regional Board pursuant to Section 13269, Water 6 Code, provided that the Forest Service reasonably implements those practices and procedures and the provisions of this agreement. However, waste discharges from land 7 management activities resulting in point source discharges, as defined by the Federal Water Pollution Act, will be subject to NPDES permit requirements, since neither the State Board 8 nor the Regional Board has authority to waive such permits.

9 AR 6029-30.23 The MAA thus specifically exempts the Forest Service from having to file reports of

10 waste discharge or from obtaining a permit or waiver. The State Board also certified a plan titled

11 “Water Quality Management for National Forest System Lands in California,” AR 3268, which

12 “obligated the USFS to incorporate Best Management Practices (BMPs) for protection of water quality

13 into land and resource management activities and to monitor their implementation and effectiveness,

14 which has been accomplished since 1992 using the BMP Evaluation Program.” AR 5840-41.

15 Plaintiffs contend that the provisions of the 1981 MAA have been “superseded by legislative

16 amendments and binding State Board nonpoint source pollution policies.” ECF No. 87 at 23. The state

17 legislature amended Porter-Cologne in 1999 to “require the [State Board] to enforce the state’s NPS

18 pollution control program.” AR 5659. The State Board responded in 2004 through its adoption of the

19 Policy for Implementation and Enforcement of the Nonpoint Source Control Program. AR 5661. That

20 policy provides that “all current and proposed NPS discharges must be regulated under [waste discharge 21 requirements], waivers of [waste discharge requirements], or a basin plan prohibition, or some combination 22 of these administrative tools.” AR 5665.

23

24 23 The MAA also included a provision stating that “nothing herein will be construed in any way as limiting the authority of 25 the State Board, o r the Regional Boards in carrying out their legal responsibilities for management, or regulation of water 2 Lahontan Regional Water Quality Control Board recognized the change, stating that “[b]ased on these

3 changes, the MAA designating the [Forest Service] as a [Water Quality Management Agency] was no

4 longer sufficient to comply with the newly adopted [Nonpoint Source] Policy.” AR 3478. The North

5 Coast Regional Board has adopted a waiver for nonpoint source discharges resulting from activities that

6 the Forest Service authorizes on its land. AR 2895, 2914-15. The State Board’s draft statewide waiver

7 that would have waived waste discharge requirements for nonpoint source discharges for certain

8 activities on National Forest system lands in California and the effort to develop a regional permitting

9 scheme would be unnecessary if the MAA still controlled, Plaintiffs contend, meaning that the Forest

10 Service is under a current obligation to file a discharge report and obtain a permit or waiver. ECF No.

11 87 at 27.

12 Regional or state efforts to evolve with the changing regulatory framework do not mean that the

13 MAA has ceased to have any legal significance. Plaintiffs have identified nothing indicating that the

14 MAA has been withdrawn or invalidated. That regional boards are in the process of developing waivers

15 or permits for activities on National Forest land is not evidence that the MAA has been withdrawn; it is 16 consistent with the MAA being in effect until superseded by a permit or waiver.24 As a result, as plainly

17 stated in the MAA, the Forest Service is not required to obtain a permit or file notices of discharge.

18 Accordingly, Plaintiffs’ motion for summary judgment on the CWA claim is DENIED, and

19 Federal Defendants and Defendant-Intervenors’ cross-motions for summary judgment are GRANTED.

20 B. Second Claim for Relief: National Forest Management Act

21 “A forest plan establishes goals and objectives for management of forest resources.” Earth

22 Island Inst. v. U.S. Forest Serv., 697 F.3d 1010, 1014 (9th Cir. 2012) (citing 16 U.S.C. § 1604(g)(1)–

23 24

25 24 The purpose of filing a notice of discharge with a regional board is to put the board on notice of a discharge so that it can 2 Rockies v. Savage, 897 F.3d 1025, 1032 (9th Cir. 2018). “Agency decisions that allegedly violated

3 NFMA . . . are reviewed under the APA.” All. for the Wild Rockies v. United States Forest Serv., 907

4 F.3d 1105, 1112 (9th Cir. 2018). See also 16 U.S.C. § 1604(i) (“Resource plans and permits, contracts,

5 and other instruments for the use and occupancy of National Forest System lands shall be consistent

6 with the land management plans.”).

7 The Forest Service completed the Stanislaus National Forest Land and Resource Management

8 Plan (“Forest Plan”) in 1991. AR 1388. The currently operative March 2017 Forest Plan Direction is

9 based on the 1991 Forest Plan, as modified by approximately two dozen Forest Plan Amendments. AR

10 1388-89. The Forest Plan contains goals that “set the standards for the future condition of the Forest.

11 Objectives and subsequent levels of direction are aimed at accomplishing these goals.” AR 1390. The

12 Forest Plan defines objectives as “planned, measurable results that respond to the general goals of the

13 Forest Plan,” and management practices as “management actions that achieve the goals and objectives

14 of the Plan.” AR 1390.

15 The Forest Goal for range is to “[m]anage livestock to utilize available forage while avoiding

16 adverse impacts on soil, vegetation, water quality, wildlife, fisheries and riparian zones.” AR 1392. The

17 Forest Goal for riparian areas is broadly to “maintain the physical, chemical and biological integrity of

18 the region’s waters” and to support the goal “to provide habitat for riparian and aquatic-dependent 19 species” under various federal laws. AR 14189-90.25

20 Forestwide standards and guidelines (“Forest Plan S&Gs”) “provide specific direction for

21 implementing the management practices throughout the Forest.” AR 1391. They apply at the project

22 level and are “an integral part of interdisciplinary planning for all projects and activities.” AR 1420. In

23 24

25 25 This Forest Go al is taken from the 2004 Sierra Nevada Forest Plan Amendment Record of Decision, which superseded the 2 short term impacts at levels which, over the long term, will facilitate meeting Forest Plan Goals and 3 Objectives.”26 AR 396. Site-specific projects, such as permits, can be challenged and must be

4 consistent with the applicable Forest Plan. Inland Empire, 88 F.3d at 757; 16 U.S.C. § 1604(i)

5 (“Resource plans and permits, contracts, and other instruments for the use and occupancy of National

6 Forest System lands shall be consistent with the land management plans.”).

7 Plaintiffs allege that the Forest Service’s actions have failed to comply with three Forest Plan 8 S&Gs: 50, 117, and 118.27 Forest Plan S&G 50 applies to range resources and provides that “[t]o

9 protect hardwood regeneration in grazing allotments, allow livestock browse on no more than 20 percent

10 of annual growth of hardwood seedlings and advanced regeneration. Modify grazing plans if hardwood

11 regeneration and recruitment needs are not being met.” AR 1439.

12 Forest Plan S&Gs 117 and 118 relate to Riparian Conservation Objective 5, which is to

13 “[p]reserve, restore, or enhance special aquatic features, such as meadows, lakes, ponds, bogs, fens, and

14 wetlands, to provide the ecological conditions and processes needed to recover or enhance the viability

15 of species that rely on these areas.” AR 1580. Forest Plan S&G 117 provides 16 Assess the hydrologic function of meadow habitats and other special aquatic features during range management analysis. Ensure that characteristics of special features are, at a 17 minimum, at Proper Functioning Condition, as defined in the appropriate Technical Reports (or their successor publications): (1) “Process for Assessing PFC” TR 1737-9 18 (1993), “PFC for Lotic Areas” USDI TR 1737-15 (1998) or (2) “PFC for Lentic Riparian- Wetland Areas” USDI TR 1737-11 (1994).

19 Id. Forest Plan S&G 118 provides

20 Prohibit or mitigate ground-disturbing activities that adversely affect hydrologic processes 21 that maintain water flow, water quality, or water temperature critical to sustaining bog and fen ecosystems and plant species that depend on these ecosystems. [Footnote 66: ground

22

23 26 This statement comes from a Forest Service document titled “Stanislaus National Forest’s Evaluation of Central Sierra Environmental Resource Center’s 2017 Meadow Condition and Grazing Utilization Report” that evaluates the data, 24 methodology, and conclusions in CSERC’s observations of the grazing allotments.

25 27 The TAC also alleges that the grazing on the BEH allotments has violated S&G 103, though Plaintiffs have abandoned this 2 m USat Dte Ar b 2e 0y 0o 4n ).d ] t Dh ue rt ih nr ge psh roo jl ed cs t i ad ne an lyti sf ii se ,d s ub ry v es yo ,i l m q au pa ,l ait ny d s dta en vd ela or pd s m. e(S asN uF reP sA to R pO roD te, c p t b. o7 g1 s; and fens from such activities as trampling by livestock, pack stock, humans, and wheeled 3 vehicles. Criteria for defining bogs and fens include, but are not limited to, presence of: (1) sphagnum moss (Spagnum spp.), (2) mosses belonging to the genus Meessia, and (3) 4 sundew (Drosera spp.)[.] Complete initial plant inventories of bogs and fens within active grazing allotments prior to re-issuing permits.

5 Id. Plaintiffs cite data from both the EIS and their own monitoring reports in support of their claims that

6 the Forest Plan S&Gs are not being met. Defendants emphasize that the Forest Service is entitled to

7 deference in the interpretation of the Forest Plan S&Gs and the actions taken in connection with their

8 implementation. Defendants also argue that the Forest Plan S&Gs further long-term goals and do not

9 serve to halt any agency action when there is any minor failure to comply. Finally, Defendants

10 challenge the data that Plaintiffs cite in support of their claims and point to recent remedial measures

11 that the Forest Service has undertaken.

12 1. The Forest Service’s Interpretation of Its Forest Plan Is Entitled to Deference

13 The parties first dispute the strength of the requirement that Forest Service actions must comply

14 with the forest plan. The core of the dispute is whether the S&Gs are firm requirements requiring strict

15 compliance or broader goals toward which short-term actions should move. In other words, the parties

16 dispute how mandatory the S&Gs really are. “While NFMA requires that the proposed site-specific

17 actions be consistent with the governing Forest Plan, the Forest Service’s interpretation and

18 implementation of its own forest plan is entitled to substantial deference.” Native Ecosystems Council v.

19 Weldon, 697 F.3d 1043, 1056 (9th Cir. 2012).

20 Plaintiffs point to 2012 regulations (which Federal Defendants call the “2012 Planning Rule”)

21 providing criteria for determining whether a project or activity is consistent with the applicable forest

22 plan, stating that a project or activity is consistent with the governing forest plan when it “complies with

23 applicable standards.” 36 C.F.R. § 219.15(d)(2). These standards also provide that consistency with the

24 governing forest plan with respect to guidelines is met when the project or activity “[c]omplies with

25 2 purpose of the applicable guidelines.” Id. § 219.15(d)(3). The 2012 Planning Rule provides that “[a]

3 standard is a mandatory constraint on project and activity decisionmaking, established to help achieve

4 or maintain the desired condition or conditions, to avoid or mitigate undesirable effects, or to meet

5 applicable legal requirements.” 36 C.F.R. § 219.7(e)(1)(3) (emphasis supplied).

6 The 2012 Planning Rule provides that forest plans that were developed, amended, or revised

7 under a prior planning regulation would be unaffected, stating that “[e]xisting plans will remain in effect

8 until revised.” 36 C.F.R. § 219.17(c). It further provides that “[n]one of the requirements of this part

9 apply to projects or activities on units with plans developed or revised under a prior planning rule until

10 the plan is revised under this part, except that projects or activities on such units must comply with the

11 consistency requirement of § 219.15 with respect to any amendments that are developed and approved

12 pursuant to this part.” Id.

13 Only one amendment to the Stanislaus Forest Plan, not at issue here, post-dates the 2012

14 Planning Rule, AR 1389 (amending SNFP for Horse Gulch Campground site, adopted Jan. 26, 2016),

15 leaving the rest of the SNFP subject to the planning rule in effect at the time adopted. See Alliance for

16 the Wild Rockies v. United States Forest Serv., 907 F.3d 1105, 1110 (9th Cir. 2018) (“Our original

17 opinion cited to 36 C.F.R. § 219.15. However, because the Payette National Forest Plan was adopted

18 pursuant to the 1982 regulations, the newer regulations, promulgated in 2012, are inapplicable. 36

19 C.F.R. § 219.17(c) (“None of the requirements of this part apply to projects or activities on units with

20 plans developed or revised under a prior planning rule until the plan is revised under this part.”)”); In re

21 Big Thorne Project, 857 F.3d 968, 974 (9th Cir. 2017) (applying “NFMA’s regulations at the time of the

22 Forest Plan”). Accordingly, the Court agrees with Federal Defendants that the governing provisions

23 here are in the 1982 Planning Rule. AR 17266 (1991 Forest Plan citing 1982 regulations); AR 14177

24 (2004 SNFPA Record of Decision, stating, “My decision conforms with the 1982 planning regulations

25 2 consistency provisions provide that “the Forest Supervisor shall ensure that, subject to valid existing

3 rights, all outstanding and future permits, contracts, cooperative agreements, and other instruments for

4 occupancy and use of affected lands are consistent with the plan.” 36 C.F.R. § 219.10(e) (1998);

5 Friends of Southeast’s Future v. Morrison, 153 F.3d 1059, 1067 (9th Cir. 1998) (quoting 36 C.F.R.

6 § 219.10(e)).

7 The Forest Service’s interpretation of its own 1982 Planning Rule is entitled to “substantial

8 deference,” and “judicial review of an agency’s interpretation of its own regulations is limited to

9 ensuring that the agency’s interpretation is not plainly erroneous or inconsistent with the regulation.”

10 Forest Guardians v. U.S. Forest Serv., 329 F.3d 1089, 1097 (9th Cir. 2003). The Forest Service views

11 the S&Gs as short-term indicators of grazing impacts that “are designed [to] keep short term impacts at

12 levels which, over the long term, will facilitate meeting Forest Plan Goals and Objectives.” AR 396.

13 “The Forest range monitoring program is geared towards assessing whether S&Gs are met across a

14 particular area over a particular time, or whether adjustments to the grazing program might be necessary

15 to effectively move towards meeting the broader Forest Goals outlined in the Forest Plan.” AR 395.

16 “[T]he focus of the Forest’s range program, which is informed by monitoring, is on meeting or moving

17 toward the broad Goals laid out in the Forest Plan, including through application of S&Gs (e.g.,

18 incorporation of applicable S&Gs in Term Grazing Permits and addressing the S&Gs in Annual

19 Operating Instructions).” Id. In other words, according to the Forest Service, general compliance with

20 short-term S&G goals will generally advance broader goals of the Forest Plan. See id. This approach is

21 consistent with the 1982 Planning Rule’s general requirement that grazing “permits . . . cooperative

22 agreements, and other instruments for occupancy and use of affected lands are consistent with the

23

24 28 Where a regulation, like the consistency provision, has been superseded, the superseded regulation applies “‘only to the extent’ it was incorporated into the relevant Forest Plan.” In re Big Thorne Project, 857 F.3d at 974 n.1 (quoting Ecology 25 Ctr. v. Castaneda , 574 F.3d 652, 657 (9th Cir. 2009)). As described in text, the 1991 Forest Plan and the 2004 Forest Plan 2 The Forest Service ensures general compliance with short-term S&G goals through two types of

3 monitoring on the BEH allotments: implementation monitoring and effectiveness monitoring. AR 8314.

4 Implementation monitoring is “short-term monitoring that is conducted to evaluate whether activities are

5 meeting Forest Plan Standards and Guidelines.” Id. It is conducted annually at “select locations on the

6 Forest.” AR 397. The USFS has limited resources—a single full-time rangeland management specialist

7 conducts most of the monitoring—and does not conduct monitoring for each parameter on each of the

8 36 grazing allotments in Stanislaus National Forest every year. Id. Instead, Forest Service employees

9 follow the Region 5 Forest Handbook to determine priority for allotment administration, using measures

10 like compliance history, resource problems, and new or emerging resource issues. Id. (citing Region 5

11 Forest Service Handbook 2209.13, section 19.1 (USDA 2011a)).

12 Effectiveness monitoring is “long-term monitoring that is conducted to determine whether the

13 Standards and Guidelines are sustaining or moving rangeland conditions toward desired conditions and

14 to establish baseline information for future planning.” AR 8314. The purpose is to “evaluate the degree

15 and rate at which prescribed management practices are meeting management objectives.” AR 399. It is

16 generally conducted every three to ten years. Id.

17 Considered as a whole, the Forest Service’s monitoring program, designed to use S&Gs as

18 stepping stones toward “desired conditions” is a reasonable interpretation of both the applicable Forest

19 Plan and the 1982 Planning Rule. With this in mind, the Court examines the Forest Service’s actions in

20 connection with each of the three challenged S&Gs.

21 2. S&G 50

22 As discussed above, S&G 50 provides “[t]o protect hardwood regeneration in grazing allotments,

23 allow livestock browse on no more than 20 percent of annual growth of hardwood seedlings and

24 advanced regeneration. Modify grazing plans if hardwood regeneration and recruitment needs are not

25 being met.” AR 1439. Plaintiffs contend that the Forest Service has violated this S&G, pointing to the 2 a. Aspen Browse

3 The Forest Service last conducted aspen browse monitoring on the BEH allotments in 2010. AR

4 399 (“Due to limited staff and resources, aspen browse is not monitored on every allotment every year.

5 Rather, Forest Service specialists may monitor aspen browse if a resource concern has been identified or

6 for upcoming project planning needs. For example, aspen browse was measured on the [BEH]

7 allotments . . . in 2010 to determine whether aspen browse standards are typically met on these

8 allotments.”). That monitoring, conducted in August and September of 2010, found that a significant

9 number of the sampled sites on the BEH allotments had aspen browse rates at or in excess of the 20%

10 threshold. AR 526-656.29 Many of the sites with total browse above 20% noted evidence of livestock.

11 AR 541 (Round Meadow, 67.8% browsed; noting signs of livestock: “fresh [hoof] prints & cow

12 patties”). Plaintiffs argue that 15 of the 28 sampled sites met or exceeded 20% browse.

13 Plaintiffs also cite CSERC’s monitoring data. The AR includes CSERC’s annual “Meadow

14 Condition and Grazing Utilization Report” for each year since 2011. See AR 657-1381. Plaintiffs assert

15 that these reports demonstrate further incidents of aspen browse exceeding 20%, and that none of this

16 has “been recorded or addressed through mitigation by the Forest Service.” ECF No. 80 at 30. The EIS

17 proposed that implementation monitoring “be conducted annually, if needed, as indicated below to

18 determine whether the Allotments are being managed in accordance with the Terms and Conditions of

19 the Grazing permits, AMPs and Annual Operating Instructions (AOIs).” AR 8499. For aspen browse

20

21 22 29 See AR 545-46 (Holding Pasture, 40% and 91.1% browsed); AR 550 (Crab Meadow, 67.8% browsed, noting that “fresh cow patties present”); AR 554 (Kerrick Corral, 34.4% browsed); AR 571 (Eagle Meadow Horse Camp, 28.9% browsed); AR 583 (Eagle Meadow, 25.6% browsed); AR 603 (Herring – Pavement End, 23.3% browsed, noting that “recent cow sign 23 ([hoof] prints & cow patties) present”); AR 612 (Fiddler’s Green, 72.2% browsed, noting “fresh cow patties present”); AR 616 (Bull Run, 35.6% browsed, noting “[r]ecent cow patties present”); AR 624 (Snow Cabin, 23.3% browsed, noting “recent 24 cow patties present”); AR 633 (Hammis West, 26.7% browsed); AR 637 (Herring Creek, 31.1% browsed); AR 645 (Cow Creek, 47.8% browsed, noting “recent cow patties present”); AR 649 (Punch Bowl, 28.9% browsed, noting presence of 25 “recent cow patti es”); AR 653 (Cow Creek, 43.3% browsed, noting “recent cow patties present”). 2 end of season.” Id. Because the Forest Service withdrew the Draft Record of Decision, the Forest 3 Service did not adopt this plan.30 Plaintiffs argue that the resulting failure to impose implementation

4 monitoring means that the acknowledged violations of the 20% browse limit remain unremedied.

5 Federal Defendants read the numbers in the EIS to show that 23 of the 36 sites exceeded 20%

6 browse but that “at least 10 of these locations had not been grazed by cattle.” ECF No. 85-1 at 29. Half

7 of the locations exceeding 20% browse did not even have cattle present, Federal Defendants argue, and

8 the Forest Service could not determine whether livestock browsing alone accounted for the exceedances 9 or whether other wildlife also contributed.31 The EIS referenced these exceedances, stating that site

10 visits to the meadows in 2004-05 and 2010-11 “indicated aspen regeneration in several areas is

11 periodically being affected by over-browsing during the late-growing season” but that “it is unknown

12 whether the majority of browse is attributable to livestock, packstock, or wildlife or a combination of all 13 browsers.”32 AR 8318. See also id. (“Aspen browsing in the project area has been noted to exceed

14 standards and guidelines in scattered locations.”). In addition, as discussed more fully below, the Forest

15 Service prepared a report assessing the data from CSERC’s monitoring and concluded that CSERC

16 failed to describe the methods used to measure aspen browse. AR 404-05. The Forest Service’s own

17

18 30 As mentioned above, in 2016, along with issuing an EIS to reauthorize grazing in the Stanislaus National Forest, the Forest Service issued a Draft Record of Decision that would have implemented an adaptive management strategy to ensure 19 compliance with the governing Forest Plan. AR 8150. But the Draft Record of Decision was withdrawn on August 29, 2016, “[i]n order to allow more interactions with stakeholders on the issues.” AR 7780. 20

31 It may be overreading the data to conclude, as Federal Defendants do, that a blank entry on the data sheets for “number of animals” implies that there were no livestock present. See ECF No. 85-1 at 29 (citing instances in the data sheets where the 21 “number of animals” field stated “none,” “non-use,” or was blank). This field is empty on every sheet other than the four instances that note that the site had no animals or was in non-use, including on sheets that also note the evidence of recent

22 cow presence. Other than the instances where this field affirmatively notes the absence of cattle, it is not clear that it offers any meaningful information. Nevertheless, there are instance of sites that exceed 20% browse that note that the area in question did not have cattle present. AR 562 (Barn Meadow Proper, 23.3% browsed (in non-use that year)); AR 566 (Barn 23 Meadow North, 20% browsed (in non-use that year)); AR 567 (Barn Meadow North, one month later, 23.3% browsed (in non-use that year)); AR 575 (Eagle Meadow Road, 23.3% browsed (no animals present)). 24 32 The EIS discussed eighteen site visits, which found six sites with aspen browsing levels exceeding 20%, though three of 25 the six included n o cattle browsing. AR 8318. 2 407. The Forest Service is conducting its own monitoring, has noted that overbrowse of aspen may not

3 be solely attributable to livestock graze, and, as discussed more fully below, is modifying grazing plans

4 where the S&G is not being met, just as S&G 50 requires. For instance, the Forest Service represents

5 that it has set up an electric fence to limit aspen browse where overbrowsing was occurring. ECF No

6 85-1 at 40 (citing AR 607-08). Given that the Forest Service has interpreted the S&Gs to require long-

7 term rather than immediate compliance, a reasonable interpretation entitled to deference, the Court

8 cannot say that the Forest Service has acted in violation of the APA with respect to aspen browse under

9 S&G 50. See Weldon, 697 F.3d at 1056 (“In determining whether [an agency decision challenged under

10 the NFMA] is arbitrary or capricious, we ‘must consider whether the decision was based on a

11 consideration of the relevant factors and whether there has been a clear error of judgment.’” (quoting

12 Morongo Band of Mission Indians v. Fed. Aviation Admin., 161 F.3d 569, 573 (9th Cir.1998))).

13 b. Woody Riparian Browse

14 Plaintiffs also point to CSERC’s monitoring data in support of their claim that there has been

15 considerable overbrowsing of willow and other riparian species in streambanks in the allotments.

16 Federal Defendants cite Forest Service monitoring data, which does not show that woody

17 riparian browse has exceeded 20% on the BEH allotments. ECF No. 85-1 at 29, 29 n.19 (citing AR 410,

18 431, 434, 494-95, 524, 525 and noting that the two locations with browse levels exceeding 20% at AR

19 8504 are not on the BEH allotments). Plaintiffs view this data as inadequate, because it views the Forest

20 Service’s method for collecting this data as inadequate. Federal Defendants describe Plaintiffs’

21 argument as a disagreement about the Forest Service’s choice of scientific methodology, the sort of

22 disagreement that goes to agency expertise. The NMFA grants the Forest Service flexibility in

23 balancing its goals and deference in achieving its objectives. In reviewing the Forest Service’s actions

24 with respect to NMFA regulations, “courts may not require a particular type of proof” but need only

25 provide “a rational connection between the facts found and the conclusions made.” In re Big Thorne 2 (9th Cir. 2008) (en banc), and Or. Nat. Res. Council Fund v. Brong, 492 F.3d 1120, 1131 (9th Cir.

3 2007)). “This rational connection can be supplied with studies or models or experts—or really any

4 legitimate evidence, so long as the agency describes a reasonable fit between its means and ends.” Id.

5 It is precisely the reasonable fit that Plaintiffs challenge. They argue that the Forest Service’s

6 use of Multiple Indictor Monitoring (“MIM”) for browse of riparian shrubs is a method ill-suited for the

7 task. ECF No. 87 at 45. The method examines a marked segment of stream selected for monitoring

8 (referred to as a designated monitoring area, or “DMA”) that extends at least 110 meters along the

9 stream. AR 1930. Examining a representative plot approximately two meters wide, the MIM method

10 involves making a visual estimate of the percentage level of browse for each woody species in a

11 particular plot. AR 1821-22. Plaintiffs contend that because the method focuses on the percentage level

12 of browse for each woody species within a plot, without regard for the number of such species within

13 the plot, thinly populated plots with unbrowsed willows would be treated with equal weight as a highly

14 populated plot with highly browsed willows. In other words, focusing on plots rather than trees could

15 result in a deeply flawed calculation, leaving the Forest Service without a way to explain adequately

16 how it is meeting the 20% standard. This is in contrast to the Forest Service protocol for aspen browse,

17 which Plaintiffs deem to be adequate because it “focuses not on an average of plots that may vary

18 considerably in measurable hardwood abundance, but rather the overall amount of browsing that has

19 occurred within a particular area.” ECF No. 80 at 32 (citing AR 399, explaining that the method counts 20 individual young aspen plants).33

21 This is nothing more than a disagreement about the Forest Service’s methodology to measure

22

23 33 Plaintiffs further point out that the technique for measuring stream alteration also uses a DMA, AR 1812-19, but register no 24 objection to its use there, because one streambank DMA is measuring the same resource as the next streambank DMA – streambank land. Plaintiffs argue that DMAs selected to measure hardwood trees, by contrast, may vary wildly in the 25 number of trees t hat are located on them, rendering the method unsupportable for measuring riparian hardwood browse. 2 method, selected based on its agency expertise, even if the court might, in the first instance, find a

3 contrary method to be a better one. See Marsh v. Oregon Nat. Res. Council, 490 U.S. 360, 378 (1989)

4 (“When specialists express conflicting views, an agency must have discretion to rely on the reasonable

5 opinions of its own qualified experts even if, as an original matter, a court might find contrary views

6 more persuasive.”). The MIM technical reference document explains that the method for measuring 7 riparian woody species use was adapted from a 1996 BLM “appearance method”34 that was considered

8 along with other methods. AR 1819-20. One of the other methods considered that is closer in substance

9 to the one that Plaintiffs advocate is the “Cole browse method,” which measures the percent of

10 individual twigs used on available shrubs, was one possibility that the Forest Service considered but

11 ultimately determined were not suitable for riparian woody species use. AR 1820 (explaining that it was

12 developed for upland shrubs, has never been used extensively on riparian shrubs, and suffered from

13 specific measurement difficulties, including an inability to differentiate use by birds from use by large

14 herbivores). The Forest Service recognizes that the MIM procedure is not a perfect one, noting that

15 there are often low numbers of woody plants and resulting small sample sizes. AR 1820-21. It also

16 cautions that the method “should not be used as a grazing use standard” but should instead “be an

17 indication of the browsing impacts within a use class range,” AR 1821, which is how the Forest Service

18 used it here. See, e.g., AR 404 (explaining that Forest Service monitoring for willow browse on (Upper)

19 Round Meadow in 2017 was 14.8%). The MIM method provides that the average use “is calculated as

20 the arithmetic average of woody species use values recorded for each plot, based upon the use category

21 or class for each species.” AR 1823. The method provides for reporting based on “class,” which is

22 broken into categories of usage: Slight (0-20%), Light (21-40%), Moderate (41-60%), Heavy (61-80%),

23 24

25 34 USDI, Bureau of Land Management. 1996b. Utilization studies and residual measurements. Interagency Technical 2 The Court agrees that a measurement protocol focusing on an average of plots rather than plants

3 has the potential for unrepresentative conclusions and that the perfect method, all else being equal,

4 would focus on plants themselves. Nevertheless, the Forest Service weighed the strengths and

5 shortcomings of the MIM method, as well as of other methods, and made the decision that the MIM

6 method was appropriate for the task. Having considered the proper factors and relied on its scientific

7 expertise to reach a reasoned conclusion, that is not a choice that the Court will second-guess.

8 3. S&G 117

9 Plaintiffs argue that the Forest Service also has violated S&G 117, which provides

10 Assess the hydrologic function of meadow habitats and other special aquatic features during range management analysis. Ensure that characteristics of special features are, at a

11 minimum, at Proper Functioning Condition, as defined in the appropriate Technical Reports (or their successor publications): (1) “Process for Assessing PFC” TR 1737-9

12 (1993), “PFC for Lotic Areas” USDI TR 1737-15 (1998) or (2) “PFC for Lentic Riparian- Wetland Areas” USDI TR 1737-11 (1994).

13

AR 1580.

14 Plaintiffs cite various statements in the EIS acknowledging some impact from grazing on special

15 aquatic features. For instance, in a summary table near the beginning of the 374-page EIS, identifying

16 existing conditions and needs for change, the EIS notes about the existing conditions that “[m]any

17 special aquatic features are functioning at risk. Livestock use may be causing trailing and trampling in

18 special aquatic features, which in some cases has resulted in hydrologic alteration, bare soil/peat and/or

19 loss of wetland vegetation.” AR 8213. The “need for change” is to “[r]educe impacts of livestock to

20 special aquatic features by improving livestock distribution and minimizing time spent near special

21 aquatic features. Update [Allotment Management Plans] to incorporate and implement an adaptive

22 strategy to improve the condition of special aquatic features that are Functioning At-Risk.” Id.

23

24

25 2 monitoring on “all known special aquatic features ([f]ens, seeps, and springs) that were identified in

3 Allotment surveys” and geographical mapping. AR 8319. That PFC monitoring found that

4 approximately half of the 44 special aquatic features assessed were properly functioning, while the other

5 half were all “Functional – At Risk.” Id. The EIS states that the “trend was not apparent” and that

6 further monitoring would be needed to determine if they were trending in an upward or downward

7 manner. Id.; id. at 8344.

8 Summarizing a 2010 assessment conducted on 16 springs, fens, and shorelines in the Bell

9 Meadow Allotment, the EIS noted that half of the special aquatic features were determined to be

10 functional – at risk, and that “the primary hydrologic issue” at those sites was “[h]oof action altering

11 flow patterns.” AR 8367. The other two allotments had similar assessments. See AR 8372 (seven of 13

12 special aquatic features surveyed in 2010 in the Eagle Meadow Allotment were functional – at risk, with

13 “[h]oof action altering flow patterns” as “the primary hydrologic issue”); AR 8377 (six of 17 special

14 aquatic features surveyed in 2010 in the Herring Creek Allotment were functional – at risk, with “[h]oof

15 action altering flow patterns” as “the primary hydrologic issue”).

16 Federal Defendants assert that more recent data since 2010 demonstrate stable conditions and

17 identify three sites that saw improved conditions between 2010 and 2015, noting that none of the

18 meadow areas on the BEH allotments examined showing deteriorating conditions. ECF No. 85-1 (citing

19 AR 433 and identifying Crab Meadow, Round Meadow, Wire Corral, and Red Rock Meadow as sites

20 with conditions that improved during the five years).

21 Part of the dispute over S&G 117 rests on what exactly the S&G requires. Federal Defendants

22 argue that the plain language of S&G 117 applies only “during range management analysis.” ECF No.

23

24 35 “PFC is a qualitative method for assessing the condition of riparian wetland areas.” AR 400. It is “a consistent approach for considering hydrology, vegetation, and soils attributes and processes to assess the condition of riparian wetland areas” 25 that “is intended to be conducted by a trained and experienced interdisciplinary team and evaluates how well the physical 2 meadow habitats and other special aquatic features” should be assessed “during range management

3 analysis,” the second component of the S&G is unbounded by a temporal limitation and requires that the

4 Forest Service “[e]nsure that characteristics of special features” are at least in PFC. ECF No. 87 at 36.

5 As mentioned above, approximately half of the special aquatic features assessed in 2010 were

6 below PFC, either “functional – at risk” or “non-functional.” Plaintiffs view this as a problem in need of

7 more serious resolution; Federal Defendants point to modest improvement on a handful of meadows and

8 argue that the special aquatic features are trending in the right direction. Exactly what the S&G means

9 when it commands the Forest Service to ensure that the characteristics of special features in PFC or

10 better is not perfectly clear, and the Forest Service asserts that the trend is in the right direction. The

11 Forest Service’s interpretation that this Forest Plan S&G is satisfied if there is an improving trend is

12 entitled to deference. The improvement may be gradual, but that does not make the Forest Service’s

13 action arbitrary or capricious.

14 4. S&G 118

15 Finally, Plaintiffs argue that the Forest Service has violated S&G 118, which provides in

16 pertinent part “[p]rohibit or mitigate ground-disturbing activities that adversely affect hydrologic

17 processes that maintain water flow, water quality, or water temperature critical to sustaining bog and fen

18 ecosystems and plant species that depend on these ecosystems.” AR 1580. It goes on to provide

19 “[d]uring project analysis, survey, map, and develop measures to protect bogs and fens from such

20 activities as trampling by livestock, pack stock, humans, and wheeled vehicles.” Id. It concludes,

21 “[c]omplete initial plant inventories of bogs and fens within active grazing allotments prior to re-issuing

22 permits.” Id.

23 Plaintiffs argue that the Forest Service has failed to comply with S&G 118 by allowing

24 continued grazing in bog and fen ecosystems that result in damage from hoofs and trampling. They cite

25 much of the same data as they do in support of their claim for S&G 117. See AR 8367 (eight of 16 2 “the primary hydrologic issue” at those sites being “[h]oof action altering flow patterns’); AR 8372

3 (seven of 13 special aquatic features surveyed in 2010 in the Eagle Meadow Allotment were functional –

4 at risk, with “[h]oof action altering flow patterns” as “the primary hydrologic issue”); AR 8377 (six of

5 17 special aquatic features surveyed in 2010 in the Herring Creek Allotment were functional – at risk,

6 with “[h]oof action altering flow patterns” as “the primary hydrologic issue”). CSERC annual

7 monitoring also found “cattle trampling altering the hydrologic functioning of seeps and fens.” ECF No.

8 75 at 28.

9 Federal Defendants state that the Forest Service has surveyed, mapped, and developed measures

10 to protect fens and bogs, such as conducting botany surveys within special aquatic features during

11 project analysis of the BEH Rangeland Allotments project, AR 8600-01, in satisfaction of the S&G’s

12 requirements. See AR 1580 (“During project analysis, survey, map, and develop measures to protect

13 bogs and fens from such activities as trampling by livestock, pack stock, humans, and wheeled

14 vehicles. . . . Complete initial plant inventories of bogs and fens within active grazing allotments prior to

15 re-issuing permits.”). It also mapped special aquatic features within the BEH project area and assessed

16 the hydrologic function of these features. AR 10499-510. Federal Defendants argue that they have

17 added mitigation features, pointing to modifications made to the grazing permits in 2017 to include a

18 “detailed map” and reference to S&G 118, with the command that pemittees protect specific identified

19 special aquatic features within the allotments. ECF No. 88 at 22 (citing AR 3, 7, 8, 11, 12). The Forest

20 Service has complied with the plain language of S&G 118, surveying and mapping during project

21 analysis as required, and mitigating ground-disturbing activities that inhibit hydrologic processes by

22 amending AOIs to identify locations of special aquatic features on the allotment, though this latter

23 component is both vaguer in the S&G and, as discussed in the discussion of S&G 117, improving at a

24 slow rate. Nevertheless, both the interpretation of the S&G and the scientific judgment driving its

25 execution are matters within the Forest Service’s discretion, and it does not appear that the Forest 2 discretion to mitigate harm when possible means that “the USFS’s alleged failure to apply these

3 provisions as [a complaining party] wishes is not a violation of the NFMA.” Conservation Cong. v.

4 United States Forest Serv., No. 18-17165, 2019 WL 2172782, at *2 (9th Cir. May 20, 2019).

5 5. Forest Service’s Disagreement with CSERC’s Data

6 The Forest Service is entitled to make its own reasonable scientific judgments so long as it has

7 considered a reasoned evaluation of the relevant factors when reaching that conclusion. Marsh, 490

8 U.S. at 378. The Forest Service has done so here in part by reviewing and analyzing CSERC’s data that

9 CSERC believes shows failure to comply with the Forest Plan S&Gs.

10 The Forest Service prepared a report evaluating CSERC’s 2017 Meadow Condition and Grazing

11 Utilization Report. AR 393 (report dated May 18, 2018); AR 657 (2017 CSERC report). The Forest

12 Service report faults the CSERC report for failing to identify the methods used to measure riparian

13 alteration and notes the Forest Service’s use of the MIM method, resulting in divergent results between

14 the CSERC report and the Forest Service monitoring data, including data taken at the same meadow just

15 a day apart. AR 404. The Forest Service report similarly faults the CSERC report for its failure to

16 describe the methods used to measure willow and aspen browse, further noting that aspen health has

17 declined in the region independent of cattle grazing. AR 404-05. The Forest Service also criticizes the

18 CSERC report for failing to describe the criteria or monitoring methods employed in determining

19 whether special aquatic features were in proper functioning condition, contrasting it with the Forest

20 Service’s use of an interdisciplinary team’s use of technical references36 to make the determinations.

21 AR 406. The Forest Service report acknowledges that it has used CSERC data to inform management

22 practices in the past and will continue to do so, but ultimately discounts CSERC’s report for using faulty

23

24 36 The Forest Service uses “A User Guide to Assessing Proper Functioning Condition and the Supporting Science for Lentic 25 Areas.” USDI T R 1737-16 (1999, revised 2003). AR 400. 2 largely inconsistent with the Forest Service’s own monitoring data for 2017 “is most likely attributable

3 to the difference in methodology employed.” AR 407. The Forest Service is entitled to rely on its own

4 expertise when it has properly considered the relevant factors. Marsh, 490 U.S. at 378 (“When

5 specialists express conflicting views, an agency must have discretion to rely on the reasonable opinions

6 of its own qualified experts even if, as an original matter, a court might find contrary views more

7 persuasive.”); Lands Council, 537 F.3d at 993 (“[O]ur law . . . requires us to defer to an agency’s

8 determination in an area involving a ‘high level of technical expertise,’” and “we are to conduct a

9 ‘particularly deferential review’ of an ‘agency’s predictive judgments about areas that are within the

10 agency’s field of discretion and expertise . . . as long as they are reasonable.’” (citations omitted)),

11 overruled on other grounds by Winter v. Nat. Res. Def. Council, Inc., 555 U.S. 7 (2008).

12 6. Forest Service’s Corrective Measures

13 Federal Defendants highlight that the Forest Service has undertaken corrective measures to try to

14 nudge the S&Gs toward compliance where appropriate. ECF No. 85-1 at 30-32, 37-38. Plaintiffs

15 repeatedly identify statements in the EIS stating that adaptive management plans should be implemented

16 to allow the Forest Service to adjust the grazing program to comply with the Forest Plan.37 For instance,

17 they point to the statement in the EIS that “[r]ecent assessments indicate that, while desired conditions

18 are being met throughout a large portion of the project area, specific locations within the project area

19 may not be meeting or moving toward desired conditions in a manner that is timely and consistent with

20 Forest Plan objectives, standards and guidelines. Gaps between existing resource conditions and desired

21 conditions indicate a need to change grazing management by updating AMPs.” AR 8199. Plaintiffs

22 also highlight language in the EIS noting that action was needed, along with the Forest Supervisor’s

23

24 37 The EIS explains that adaptive management “uses monitoring to determine if actions prescribed were followed, and adjusts 25 management if ch anges are needed. An adaptive strategy is particularly suited for dealing with management issues involving 2 that specific locations within the project area may not be meeting or moving toward desired conditions

3 in a manner that is timely and consistent with the Forest Plan. There is a need to change grazing

4 management by implementing an adaptive management system and resource conservation measures that

5 more effectively move resource conditions toward desired conditions in a manner that is timely and 6 consistent with the Forest Plan.”38 AR 8165.

7 Plaintiffs view the current monitoring framework as too infrequent to ensure compliance and

8 argue that the failure to adopt the adaptive management plan in the EIS leaves the Forest Service on the

9 same path that has resulted in “decades of non-compliance with these standards.” ECF No. 87 at 41

10 (citing AR 8210). More importantly, Plaintiffs contend that failure to complete the NEPA process

11 leaves the Forest Service without the authority to impose meaningful adaptive-management-triggered

12 restrictions like adding recommendations or instructions in an AOI, because such actions require the

13 completion of the NEPA process. Id. (citing AR 18392 (FSH Section 96.2: “Adaptive management

14 options that would be activated if the authorized activity is not achieving the anticipated objectives must

15 be specified in the project-level decision.”); AR 18385 (FSH Section 92.23b(3): “As circumstances

16 where changes in conditions warrant implementation of a management option that has not been provided

17 for in the NEPA analysis, or when the predicted effects of implementation are determined to be greater

18 than the effects originally predicted, a supplemental or new NEPA analysis and NEPA-based decision is

19 needed.”); AR 8239 (EIS statement that under current management, adoptive management options are

20 limited to the terms of the grazing permit)). In Plaintiffs’ view, the adaptive management options that

21 the Forest Service retains in the absence of the completion of the NEPA process is too toothless to

22 comply with the S&Gs. ECF No. 87 at 37-42.

23

24 38 The Draft Record of Decision also proposed an action that included “[a]uthorization of continued livestock grazing within the permitted grazing area will occur the year following this decision with updates to AMPs and term grazing permits that 25 would establish t he specified livestock numbers, season of use, utilization standards, allotment infrastructure and 2 2017 to provide additional information, including a “detailed map” of special aquatic features on the

3 allotment and reference to S&G 118 with the command that permittees protect specific identified special

4 aquatic features within the allotments. ECF No. 88 at 22 (citing AR 3, 7, 8, 11, 12). The 2017

5 modifications also specify which S&Gs apply to ground-disturbing activities and which special aquatic

6 features are on the allotments. Id. The Forest Service has also recently modified AOIs. AOIs since

7 2016 have included a new section entitled “Other Management Recommendations/Instructions” or

8 “Other Management Considerations.” AR 78, 85, 92 100, 108, 115, 118, 121, 133. These new sections

9 include mitigation measures, such as installing an electric fence, herding away from Wire Corral,

10 minimizing use on Coyote Meadow, and installing a water trough on Bluff Meadow. AR 121, 133. The

11 Forest Service also represents that the Forest Service set up an electric fence to protect aspen where

12 excessive aspen browse was an issue. ECF No. 85-1 at 40 (citing AR 607-08 (aspen browse data sheets

13 noting that “clump is surrounded by electric fencing”)). The Forest Service characterizes these steps as

14 improvements to the range management process since 2016. They argue that the Forest Service was

15 already complying with the Forest Plan and that these more recent efforts have only improved the range

16 monitoring program. ECF No. 88 at 24. Though the Forest Service did not adopt the Draft Record of

17 Decision, it did incorporate adaptive management principles into its administration of grazing on the

18 BEH allotments.

19 Plaintiffs dismiss these “[r]andom mitigation measures” as an inadequate substitute for “a

20 coherent strategy and regulatory commitment to limit grazing when adverse impacts occur.” ECF No.

21 87 at 42. An adaptive management system can come in many varieties and need not follow the exact

22 contours of the adaptive management proposed in the Draft Record of Decision. The Forest Service is

23 entitled to craft the adaptive management program that it deems appropriate. The Forest Service’s

24 actions may not represent Plaintiffs’ preferred course, but as long as it has reasonably interpreted its

25 Forest Plan and used its scientific judgment while considering the proper factors, the decision belongs to 2 The alternatives outlined in the EIS are not the measuring stick against which all adaptive

3 management activities must be compared. Even in the absence of a completed NEPA process, the

4 Forest Service retains the power to impose directions and limitations on grazing in the BEH allotments.

5 The corrective measures outlined in the EIS may be swifter and more forceful than the measures the

6 Forest Service has undertaken so far. But if the Forest Service has determined in its expertise that its

7 Forest Plan requires movement toward long-term goals in the S&Gs and that it is on the path to

8 achieving those goals, that judgment is entitled to deference. Plaintiffs have not demonstrated that the

9 Forest Service’s judgment should be disturbed. Forest Guardians, 329 F.3d at 1097-98 (holding that in

10 light of the deferential standard of review of the Forest Service’s interpretation of its forest plan and in

11 light of the federal requirement to consider the use of National Forest lands for grazing of livestock and

12 to develop AMPs in consultation with permittees, lessees, and landowners, a Forest Service decision to

13 reduce grazing over three-year period rather than doing so immediately was not arbitrary, capricious, or

14 a violation of law); id. at 1099-1100 (“An agency’s actions need not be perfect; we may only set aside

15 decisions that have no basis in fact, and not those with which we disagree. . . . Thus, even if we were to

16 conclude that the Service could develop a better system of predicting wild ungulate use, or even

17 preventing overgrazing, we are not permitted to substitute our judgment for the agency’s.”); Theodore

18 Roosevelt Conservation P’ship v. Salazar, 616 F.3d 497, 517 (D.C. Cir. 2010) (holding in a NEPA

19 challenge that “[a]llowing adaptable mitigation measures is a responsible decision in light of the

20 inherent uncertainty of environmental impacts” that “is certainly not arbitrary or capricious”).

21 In sum, Federal Defendants argue that the Forest Service’s range monitoring program reasonably

22 assesses compliance with the S&Gs, that the Forest Service has in fact complied with S&Gs 50, 117,

23 and 118, and that these methods and conclusions are entitled to deference because a court’s “highest

24 deference is owed to the Forest Service’s technical analyses and judgments within its area of expertise.”

25 League Of Wilderness Defs. Blue Mountains Biodiversity Project v. Allen, 615 F.3d 1122, 1131 (9th Cir. 2 Council, Inc., 555 U.S. 7 (2008)). The Court agrees.

3 Accordingly, Plaintiffs’ motion for summary judgment on the NMFA claim is DENIED, and

4 Federal Defendants and Defendant-Intervenors’ cross-motions for summary judgment are GRANTED.

5 VI. CONCLUSION AND ORDER

6 The Court is sympathetic to the concerns that sparked this lawsuit. Technical violations of water

7 quality standards persist, and the process in place to address them is moving at a glacial pace. Yet, the

8 structure of the regulatory regimes at issue in this case grants state water quality regulators great

9 flexibility in their dealings with the Forest Service, which is, in turn, working cooperatively with

10 regulators to address water quality issues. Likewise, the Forest Service is entitled to deference in its

11 achievement of compliance with the various goals set forth in the Forest Plan. Accordingly, and for the

12 reasons set forth above Plaintiffs’ motion for summary judgment is DENIED and Defendants’ cross

13 motions are GRANTED.

14 Within ten days of electronic service of this Memorandum Decision and Order, Federal

15 Defendants shall submit a proposed form of judgment consistent with the rulings herein.

16 IT IS SO ORDERED.

17 Dated: August 6, 2019 /s/ Lawrence J. O’Neill _____

UNITED STATES CHIEF DISTRICT JUDGE

18

19

20

21

22

23

24

25