FOR PUBLICATION

UNITED STATES COURT OF APPEALS

FOR THE NINTH CIRCUIT

STATE OF MONTANA, No. 23-3050 D.C. No. Plaintiff - Appellant, 6:16-cv-00035- v.

DLC

TALEN MONTANA, LLC;

NORTHWESTERN

CORPORATION; UNITED OPINION

STATES OF AMERICA; UNITED

STATES FOREST SERVICE;

UNITED STATES BUREAU OF

RECLAMATION; UNITED

STATES BUREAU OF LAND

MANAGEMENT,

Defendants - Appellees.

STATE OF MONTANA, No. 23-3353 D.C. No. Plaintiff - Appellee, 6:16-cv-00035- v. DLC

TALEN MONTANA, LLC;

NORTHWESTERN

CORPORATION,

Defendants - Appellants, 2 STATE OF MONTANA V. TALEN MONTANA, LLC

and

UNITED STATES OF AMERICA,

UNITED STATES FOREST

SERVICE, UNITED STATES

BUREAU OF RECLAMATION,

UNITED STATES BUREAU OF

LAND MANAGEMENT,

Defendants.

Appeal from the United States District Court for the District of Montana Dana L. Christensen, District Judge, Presiding

Argued and Submitted January 15, 2025 San Francisco, California

Filed March 4, 2025

Before: Holly A. Thomas, Salvador Mendoza, Jr., and Ana de Alba, Circuit Judges.

Opinion by Judge de Alba

STATE OF MONTANA V. TALEN MONTANA, LLC 3

SUMMARY *

Title to Riverbeds

In an appeal by the State of Montana and a cross-appeal by owners and operators of hydroelectric dams, the panel affirmed the district court’s judgment (1) quieting title to the United States for the riverbeds underlying four designated “Segments” of rivers within Montana’s borders, and (2) quieting title to Montana for the riverbeds within the Sun River to Black Eagle Falls Segment of the Missouri River. Whether Montana or the United States holds title depends on whether the rivers were “navigable in fact” at the time of Montana’s statehood in 1889. Upon statehood, the State gained title within its borders to the beds of waters then navigable, while the United States retained title to riverbeds underlying non-navigable rivers. Applying the “navigability in fact” test for determining riverbed title, as clarified in PPL Montana, LLC v. Montana, 565 U.S. 576 (2012), the district court found only one Segment—the Sun River to Black Eagle Falls Segment—to be navigable in fact. Addressing the State of Montana’s appeal, the panel held that the district court correctly applied the PPL framework to the evidence and did not violate any PPL mandate in quieting title to the United States for the riverbeds underlying four designated Segments of rivers within Montana’s borders. The panel affirmed the district

* This summary constitutes no part of the opinion of the court. It has been prepared by court staff for the convenience of the reader. 4 STATE OF MONTANA V. TALEN MONTANA, LLC

court’s analysis of each Segment in their entirety and held that it correctly identified and analyzed each segment under PPL, which requires that navigability for title must be determined on a segment-by-segment basis. The panel rejected Montana’s argument that evidence of “actual use” of the rivers, by itself, establishes navigability. The panel also rejected Montana’s arguments challenging the district court’s factual findings and conclusions that the four Segments were not navigable. On cross-appeal, the panel rejected Talen Montana, LLC and Northwestern Corporation’s argument that the district court should not have considered the navigability of the Sun River to Black Eagle Falls Segment because it is part of the “17-mile Great Falls reach” that PPL held was not navigable. The panel held that the district court’s review and ruling of navigability of the Sun River to Black Eagle Falls Segment was consistent with the mandate from

PPL.

COUNSEL

John E. Bloomquist (argued) and Betsy R. Story, Parsons Behle & Latimer, Helena, Montana; Brent A. Mead, Assistant Solicitor General; Patrick M. Risken, Assistant Attorneys General; Austin Knudsen, Montana Attorney General; Montana Attorney General’s Office, Helena, Montana; Brian K. Gallik, Gallik & Bremer PC, Bozeman, Montana; Emily Jones, Jones Law Firm PLLC, Billings, Montana; for Plaintiff-Appellant. Robert L. Sterup (argued), Brown Law Firm, Billings, Montana; Christopher Anderson (argued), David W. Gehlert, J. Scott Thomas, and John L. Smeltzer, Attorneys;

STATE OF MONTANA V. TALEN MONTANA, LLC 5

Todd Kim, Assistant Attorney General; Environment & Natural Resources Division, United States Department of Justice, Washington, D.C.; Kyle A. Gray, Holland & Hart LLP, Billings, Montana; Brian B. Bell and B. Andrew Brown, Dorsey & Whitney LLP, Minneapolis, Minnesota; Stephen D. Bell, Dorsey & Whitney LLP, Missoula, Montana; for Defendants-Appellees.

OPINION

DE ALBA, Circuit Judge:

Since embarking from Montana state court over 20 years ago, this riverbed title dispute has traveled from the Montana Supreme Court to the United States Supreme Court, then back to Montana and into the United States District Court for the District of Montana for a 10-day bench trial. It has now landed on our docket for review. The State of Montana appeals the district court’s ruling that quieted title to the United States for the riverbeds underlying four designated “Segments” of rivers within Montana’s borders. The parties’ experts designated these Segments as the Big Belt Mountains Segment and the Big Falls to Belt Creek Segment of the Missouri River, the Eddy Segment of the Clark Fork River, and the Headwaters/West Yellowstone Basin Segment of the Madison River. 1 Montana argues that the district court misapplied the “navigability in fact” test for determining

1 The district court quieted title to riverbeds within several other Segments to the United States, but Montana only challenges its ruling as to these four. 6 STATE OF MONTANA V. TALEN MONTANA, LLC

riverbed title, which the Supreme Court reaffirmed and clarified in this case in PPL Montana, LLC v. Montana, 565 U.S. 576 (2012). On cross-appeal, Talen Montana, LLC (formerly PPL Montana, LLC) and NorthWestern Corporation, seek to reverse the district court’s quieting of title to Montana for the riverbeds within the only river segment Montana won at trial, known as the Sun River to Black Eagle Falls Segment of the Missouri River. Talen and NorthWestern do not argue over navigability. Instead, they contend that this Segment is part of the “17-mile Great Falls reach,” the entirety of which they argue the Supreme Court already ruled is not navigable in PPL and seek reversal under the mandate rule. We hold that the district court correctly applied the PPL framework to the evidence and did not violate any PPL mandate. We therefore affirm the district court’s judgment.

I. Background

At its core, this case is a land ownership and rent dispute between Montana, the United States, and Talen and NorthWestern. Talen and NorthWestern own and operate hydroelectric dams located along portions of the Missouri, Clark Fork, and Madison Rivers within Montana’s borders. The parties’ dispute turns on who holds title to underlying riverbeds where the dams are situated: Montana or the United States. Whether Montana or the United States holds title depends on whether the rivers were “navigable in fact” at the time of Montana’s statehood in 1889. Talen and NorthWestern owe rent to the title holder. For many decades before this case started in 2003, PPL Montana had been paying rent to the United States without any objection

STATE OF MONTANA V. TALEN MONTANA, LLC 7

from Montana under the belief that the federal government held title. See PPL, 565 U.S. at 586–87. The Supreme Court recounted the history of this case in its decision in PPL, which we need not repeat. See id. at 586–89. In reversing Montana’s initial win in its state courts, the Supreme Court reiterated and clarified the legal standards and requirements of title navigability that the district court followed as a “roadmap” and that govern our review. See id. at 589–602. A. The PPL Framework The Court in PPL explained that the “navigability in fact” test for riverbed title is rooted in the equal footing doctrine. See 565 U.S. at 590–92. The equal footing doctrine provides that “[u]pon statehood, the State gains title within its borders to the beds of waters then navigable[,]” while the United States retains title to riverbeds underlying non-navigable rivers. Id. at 591. Rivers are “navigable in fact” when they are “used, or are susceptible of being used, in their ordinary condition, as highways for commerce, over which trade and travel are or may be conducted in the customary modes of trade and travel on water.” Id. at 592 (quoting The Daniel Ball, 77 U.S. (10 Wall.) 557, 563 (1870)). The Court emphasized, as we have, that the “navigability in fact” test for deciding riverbed title is legally distinct from determining navigability for purposes of federal regulatory authority. See id. at 592, 598; see also Boone v. United States, 944 F.2d 1489, 1499 (9th Cir. 1991) (“Cases interpreting navigability cannot be simply lumped into one basket . . . any reliance upon judicial precedent must be predicated upon careful appraisal of the purpose for which the concept of navigability was invoked in a particular case.” (quotation 8 STATE OF MONTANA V. TALEN MONTANA, LLC

marks omitted)). The main difference is that, for determining title to a given riverbed, we focus on the “natural and ordinary condition of the water” at the time the State became a state. PPL, 565 U.S. at 592 (quotation marks omitted). Most relevant to this appeal, PPL reaffirmed the requirement that courts consider a river’s navigability on a “segment-by-segment” basis. Id. at 593–94 (first citing United States v. Utah, 283 U.S. 64, 77 (1931) and then citing Brewer-Elliott Oil & Gas Co. v. United States, 260 U.S. 77, 85 (1922)). Segmentation accounts for the practical reality that “[p]hysical conditions that affect navigability often vary significantly over the length of a river.” Id. at 595. To identify a particular river “segment” where a disputed area lies, courts must look to the physical characteristics of a river. See id. (“[S]hifts in physical conditions provide a means to determine appropriate start points and end points for the segment in question.”). Segments should be “both discrete, as defined by physical features characteristic of navigability or nonnavigability, and substantial, as a matter of administrability for title purposes.” Id. at 597. Courts must apply the segment-by- segment approach “sensibly” in order to “determine[] precisely” the “exact point at which navigability may be deemed to end.” Utah, 283 U.S. at 90. The Court also described the effect of land portages— where a river traveler must exit the water and proceed on land, either abandoning, carrying, or pulling the boat—to circumvent part of the river that the boat cannot move through. See PPL, 565 U.S. at 597–99. Portages generally “demonstrate[] the need to bypass the river segment, all because that part of the river is nonnavigable.” Id. at 597.

STATE OF MONTANA V. TALEN MONTANA, LLC 9

Portages therefore “may defeat navigability for title purposes.” Id. at 599. Some “present-day” or post-statehood evidence may be relevant to title navigability, but such evidence must “be confined to that which shows the river could sustain the kinds of commercial use that, as a realistic matter, might have occurred at the time of statehood.” Id. at 600. The party who relies on “present-day evidence” must show two things: “(1) the watercraft are meaningfully similar to those in customary use for trade and travel at the time of statehood; and (2) the river’s poststatehood condition is not materially different from its physical condition at statehood.” Id. at 601. The Supreme Court remanded the case to the Montana state courts to determine navigability of the disputed rivers consistent with its opinion. See id. at 600, 605. Eventually, the case was removed to federal court and the United States was later joined as a necessary party. B. Talen and NorthWestern’s Motion to Dismiss In PPL, the Supreme Court used the “Great Falls reach” of the Missouri River as an example to highlight the Montana state courts’ errors and misapplication of title navigability law. See id. at 586, 596–97, 599. The Court described the Great Falls reach as “a 17–mile stretch that begins somewhat above the head of Black Eagle Falls,” that drops “over 400 feet within 10 miles from the first rapid to the foot of Great Falls,” and contains “distinct drops including five waterfalls and continuous rapids in between.” Id. at 584, 597. It noted that five of the seven Missouri River dams at issue in the case are located along the reach, including the Black Eagle Falls Dam. Id. at 584, 586. And it found that when Meriwether Lewis and 10 STATE OF MONTANA V. TALEN MONTANA, LLC

William Clark explored the territory in 1805, they “transported their supplies and some small canoes about 18 miles over land” to circumvent the Great Falls “and their surrounding reach of river.” Id. at 583–84, 597. Because Lewis and Clark and anyone else seeking to traverse the river had to portage around the falls, the Court held that “the 17–mile Great Falls reach, at least from the head of the first waterfall to the foot of the last, is not navigable for purposes of riverbed title.” Id. at 599. But it did not identify clear start or end points for the “17–mile Great Falls reach” or analyze navigability in any more detail. Despite the Supreme Court’s discussion regarding the Great Falls reach, Montana’s complaint on remand still claimed title to the areas where the five dams along the reach are located. Talen and NorthWestern moved to dismiss Montana’s claims to these areas based on PPL’s ruling as to the Great Falls reach. After carefully analyzing the PPL decision, the district court granted the motions in part and denied them in part. It held that the Supreme Court decided that “at least from the head of the first waterfall [Black Eagle Falls] to the foot of the last [the Great Falls],” the Missouri River was not navigable. It therefore granted the motions to the “roughly 8.2 river-mile stretch demarcated as the head of the Black Eagle Falls to the foot of the Great Falls.” But the district court otherwise denied the motions, finding that “the mandate should only pertain to the caveat explicitly carved into the Supreme Court’s conclusion regarding the Great Falls reach.” The district court’s partial denial of the motions to dismiss is the subject of Talen’s and NorthWestern’s cross- appeal.

STATE OF MONTANA V. TALEN MONTANA, LLC 11

C. The Bench Trial and Relevant Rulings The parties agreed to bifurcate the case into two phases. Phase I addressed Montana’s riverbed title claims and Talen and NorthWestern’s liability and defenses, and Phase II would concern damages. In January 2022, the district court held a 10-day bench trial that included hundreds of exhibits and testimony from 15 expert witnesses. On August 25, 2023, the district court issued 77 pages of detailed findings of fact and conclusions of law that quieted title for the disputed riverbed lands. The district court first divided the three rivers into 17 different “Relevant Segments” based on “physical conditions affecting navigability” that the parties’ expert geomorphologists identified. We emphasize here that Montana’s experts largely agreed with Talen’s and NorthWestern’s experts on the appropriate Segment boundaries based on critical navigability factors like water depth, the gradient of the terrain, velocity of the water, and the presence of obstructions like rapids or waterfalls, among others. No party challenges the Segment boundaries on appeal. Within the Relevant Segments, the court identified seven different “Disputed Reaches.” The Disputed Reaches are smaller sub-sections of the Relevant Segments and are based on regulatory boundaries that the Federal Energy Regulatory Commission (FERC) established in licensing the dams. The FERC boundaries have nothing to do with navigability or any physical characteristics of the 12 STATE OF MONTANA V. TALEN MONTANA, LLC

river and, of course, were established long after 1889. See 42 U.S.C. § 7134; 18 C.F.R. § 4.41(b), (h). 2 Next, the district court identified the “customary modes of trade and travel” at the time of Montana’s statehood. Based on the parties’ evidence, it concluded that the relevant statehood-era watercraft included the upland steamboat and smaller watercraft, namely “dugout canoes, bull boats, skin canoes, bark canoes, bateaus, mackinaws, [and] keelboats.” As to the findings of “navigability in fact,” the district court methodically analyzed each river, segment by segment. It found the two Relevant Segments of the Missouri River were not navigable at Montana’s statehood. The Big Belt Mountains Segment lacked persuasive evidence showing actual use of the river for trade or travel, and was overall too shallow, and contained too many rapids (such as the Beartooth Rapids) with steep gradients and fast-moving water to be susceptible to navigation in its “natural and ordinary” conditions. As to the Big Falls to Belt Creek Segment, there was no evidence of actual use, and the presence of statehood-era waterfalls, bedrock shoals, continuous rapids, steep gradients, and shallow depths made this segment not susceptible to navigation. The evidence led the district court similarly to find that neither the Eddy Segment of the Clark Fork River nor the Headwaters/West Yellowstone Basin Segment of the Madison River were navigable. The district court found only one Segment to be navigable in fact: the Sun River to Black Eagle Falls

2 Montana argues it is specifically claiming title to—and back rent for—the riverbed portions underlying the Disputed Reaches.

STATE OF MONTANA V. TALEN MONTANA, LLC 13

Segment, which lies just upstream of the head of Black Eagle Falls. There was sufficient evidence of actual use of this Segment, and sufficient depths and modest gradients made it susceptible to trade and travel at statehood. No party challenges the district court’s navigability finding as to this Segment. The focus on appeal is whether the rivers’ physical qualities defeat navigability. No party disputes the district court’s findings about any river’s physical qualities (such as depth or water velocity) on appeal, but their experts certainly disagreed at trial. Two expert opinions are worth noting here. The first is Montana’s geomorphology expert Dr. Andrew Wilcox, who opined that the Big Falls to Belt Creek Segment could be navigated at statehood based on his findings of depth, gradient, and water velocity. Dr. Wilcox conceded, however, that his calculations were “back of the envelope” extrapolations based on a river gage about 45 miles away from the Segment. As Dr. Wilcox failed to account for significant physical differences between the Segment and his baseline river gage, the district court found his conclusions to be “unpersuasive” and “unreliable.” Dr. Wilcox’s conclusions as to the Eddy Segment were similarly defective. The second notable opinion was from Montana’s expert boatman, Jason Cajune. Mr. Cajune is an elite river guide whose testimony the district court found “to be informative and reliable on many of the issues in this case.” But the district court ultimately found Mr. Cajune’s conclusions about legal navigability to be unpersuasive because his own 14 STATE OF MONTANA V. TALEN MONTANA, LLC

skills did not reliably translate to a river’s historic susceptibility to commercial navigation. 3

II. Standards of Review

A. Title Navigability A district court’s findings of fact after a bench trial are reviewed for clear error. Fed. R. Civ. P. 52(a)(6). Legal conclusions are reviewed de novo. Chaudhry v. Aragón, 68 F.4th 1161, 1171 (9th Cir. 2023). Under the clear error standard, we reverse “only if the district court’s findings are . . . illogical, implausible, or without support in inferences from the record.” Id. (alteration in original). We must have a “definite and firm conviction that a mistake has been committed” to justify reversal. Anderson v. Bessemer City, 470 U.S. 564, 573 (1985). A district court’s plausible account of the evidence is not reversible even if we are “convinced that . . . [we] would have weighed the evidence differently.” Id. at 574. We “must be especially reluctant to set aside a finding based on the trial judge’s evaluation of conflicting lay or expert oral testimony.” Beech Aircraft Corp. v. United States, 51 F.3d 834, 838 (9th Cir. 1995). An appellate court oversteps Rule 52(a) “if it undertakes to duplicate the role of the lower court” and decides factual issues. Anderson, 470 U.S. at 573. Rule 52(a) does not limit our power to “correct errors of law, including those that may infect a so-called mixed finding of law and fact, or a

3 Montana argues that the district court imposed a “skill of operator” standard into its findings based on its discussion about Mr. Cajune’s expertise. We disagree. The record is clear that the district court discussed Mr. Cajune’s skills relative to frontier Montanans while evaluating the weight of Mr. Cajune’s opinions, not as a new “element” for navigability.

STATE OF MONTANA V. TALEN MONTANA, LLC 15

finding of fact that is predicated on a misunderstanding of the governing rule of law.” Bose Corp. v. Consumers Union of U.S., Inc., 466 U.S. 485, 501 (1984) (citations omitted). “[A] reviewing court should try to break [a mixed question of law and fact] into its separate factual and legal parts, reviewing each according to the appropriate legal standard.” Google LLC v. Oracle Am., Inc., 593 U.S. 1, 24 (2021). “But when a question can be reduced no further, . . . ‘the standard of review for a mixed question all depends[] on whether answering it entails primarily legal or factual work.’” Id. (quoting U.S. Bank Nat’l Ass’n v. Vill. at Lakeridge, LLC, 583 U.S. 387, 396 (2018)). Montana strains to argue for de novo review of the district court’s navigability findings, but to no avail. “Each determination as to navigability must stand on its own facts.” Utah, 283 U.S. at 87 (emphasis added). Navigability-in-fact, as the name implies, is highly fact- intensive and necessarily “immerse[s] courts in case- specific [or here, river-specific] factual issues—compelling them to marshal and weigh evidence, make credibility judgments, and otherwise address . . . ‘multifarious, fleeting, special, narrow facts that utterly resist generalization,’” all which calls for clear error review. Lakeridge, 583 U.S. at 396 (quoting Pierce v. Underwood, 487 U.S. 552, 561–62 (1988)). That we apply “a legal standard to a particular set of facts” does not transform the fact-heavy inquiry into a legal one demanding de novo review. Miller v. Thane Int’l, Inc., 615 F.3d 1095, 1104 (9th Cir. 2010) (quoting TSC Indus., Inc. v. Northway, Inc., 426 U.S. 438, 450 (1976)). Wading deep into the evidentiary waters of this case to decide facts of navigability is improperly duplicative of the district court’s already thorough review. See Fed. R. Civ. 16 STATE OF MONTANA V. TALEN MONTANA, LLC

P. 52(a); Anderson, 470 U.S. at 573. Indeed, all of Montana’s arguments eventually require crediting some evidence and expert opinions over others. Tellingly, Montana never addresses Rule 52 and even concedes that the “evidentiary basis” upon which navigability is decided is a factual issue. We therefore review the district court’s findings of navigability for each Segment for clear error. Accord N.C. Dep’t of Admin. v. Alcoa Power Generating, Inc., 853 F.3d 140, 151–52 (4th Cir. 2017) (reviewing the district court’s navigability findings for clear error). B. The Mandate Rule Whether a district court followed the mandate of a higher court is a question of law reviewed de novo. See United States v. Kellington, 217 F.3d 1084, 1092 (9th Cir. 2000) (citations omitted).

III. Discussion

To claim title, Montana must show that the riverbeds underlying the Big Belt Mountains, Big Falls to Belt Creek, Eddy, and Headwaters/Yellowstone Basin Segments were “navigable in fact” at the time of its statehood in 1889. See PPL, 565 U.S. at 592. To prove navigability in fact, Montana must demonstrate that those segments are “used, or are susceptible of being used, in their ordinary condition, as highways for commerce, over which trade and travel are or may be conducted in the customary modes of trade and travel on water.” Id. (citation omitted). “[T]he evidence must be confined to that which shows the river could sustain the kinds of commercial use that, as a realistic matter, might have occurred at the time of statehood.” Id. at 600.

STATE OF MONTANA V. TALEN MONTANA, LLC 17

Evidence of “actual use of streams, and especially of extensive and continued use for commercial purposes” is the most persuasive evidence of navigability. Utah, 283 U.S. at 82. Actual use evidence may include the use of “boats of various sorts” for exploration, recreation, transportation, travel, and surveying. Id.; see also United States v. Holt State Bank, 270 U.S. 49, 57 (1926). Evidence of log drives may also demonstrate actual use “when joined with the other facts of the case.” Oregon v. Riverfront Prot. Ass’n, 672 F.2d 792, 794 (9th Cir. 1982). But the “mere fact that logs, poles, and rafts are floated down a stream occasionally and in times of high water does not make it a navigable river.” United States v. Rio Grande Dam & Irrigation Co., 174 U.S. 690, 698 (1899). Susceptibility of a river to “use by the public for purposes of transportation and commerce” is the “true criterion of the navigability of a river.” Utah, 283 U.S. at 83 (quoting Econ. Light & Power Co. v. United States, 256 U.S. 113, 122–23 (1921)); see also The Montello, 87 U.S. (20 Wall.) 430, 441 (1874). The “vital and essential point is whether the natural navigation of the river is such that it affords a channel for useful commerce.” Montello, 87 U.S. at 443 (emphasis added). Susceptibility evidence can include “physical characteristics [of the segment] and experimentation as well as by the uses to which the streams have been put,” including evidence of impediments like rapids, riffles, or speed of the current. Utah, 283 U.S. at 83–85. Difficulties in navigation do not necessarily defeat navigability. Navigability “does not depend . . . upon the difficulties attending navigation, but upon the fact whether the river in its natural state is such that it affords a channel for useful commerce.” Brewer-Elliott, 260 U.S. at 86 18 STATE OF MONTANA V. TALEN MONTANA, LLC

(citations omitted); see also Riverfront Protection, 672 F.2d at 795 (“[U]se of the river need not be without difficulty, extensive, or long and continuous.”). Nor does navigability “depend upon the mode by which commerce is conducted upon it, whether by steamers, sailing vessels or flat boats,” so long as the evidence of the watercraft considered is “meaningfully similar to those in customary use for trade and travel at the time of statehood.” Brewer-Elliott, 260 U.S. at 86; PPL, 565 U.S. at 601. Still, “it is not . . . every small creek in which a fishing skiff or gunning canoe can be made to float at high water which is deemed navigable . . . it must be generally and commonly useful to some purpose of trade or agriculture.” Montello, 87 U.S. at 442 (internal quotations omitted). In addition to its disagreements with the district court’s factual findings, Montana raises two legal issues: first, that the district court erred by analyzing the Relevant Segments in their entirety, rather than narrowing its review to the Disputed Reaches only; and second, that the district court did not properly credit Montana’s “actual use” evidence. We are not persuaded. A. Application of the PPL Segment-by-Segment Analysis The Supreme Court was clear that navigability for title must be determined on a segment-by-segment basis. See PPL, 565 U.S. at 594 (“The segment-by-segment approach to navigability for title is well settled, and it should not be disregarded.”). The Segments must be “discrete, as defined by physical features characteristic of navigability or nonnavigability, and substantial, as a matter of administrability for title purposes.” Id. at 597 (emphasis added). The inquiry must focus on the “natural and

STATE OF MONTANA V. TALEN MONTANA, LLC 19

ordinary condition” of the river segments at the time of Montana’s statehood, and post-statehood or modern-day evidence is only acceptable if Montana shows that “the river’s poststatehood condition is not materially different from its physical condition at statehood.” Id. at 601. It is worth reiterating that Montana itself proposed the Segment boundaries that the district court accepted, with some small adjustments. During trial, Montana presented evidence about the navigability of the entirety of each Segment and focused on the ability of smaller watercraft— like mackinaws and canoes—to navigate those segments. But now on appeal, Montana urges us to disregard their own Segments and look only at the narrower “Disputed Reaches” for navigability. Arguable waiver aside, Montana’s assertion that segment-by-segment review applies only to the Disputed Reaches is meritless. PPL is clear that physical characteristics of rivers determine the “start points and end points” of a given segment. Id. at 595. The Relevant Segments established at trial are based on the rivers’ physical qualities at statehood. By contrast, as Montana acknowledged at oral argument, the Disputed Reaches are defined by the FERC project boundaries, which have nothing to do with any physical characteristic of the rivers or their navigability at statehood. Montana would no doubt prefer us to focus on narrower parts of the river given that many of the major barriers to navigation in this case lie outside the Disputed Reaches, but within the Relevant Segments. But that highlights the problem with Montana’s post-trial approach: focusing on the Disputed Reaches ignores the physical realities of the surrounding river, thus directly contradicting PPL and undermining the segment- by-segment approach. See 565 U.S. at 594 (discussing a 20 STATE OF MONTANA V. TALEN MONTANA, LLC

prior case that found one segment of river was not navigable “even though a segment of the river that began further downstream was navigable.”). Indeed, under Montana’s approach, a state plaintiff could almost always obtain title to a riverbed with careful pleading and naming conventions, regardless of the rivers’ physical qualities that affect navigability. There is no support in law or the record that allows Montana to change course and narrow our review to its now-preferred portions of the rivers in question. We affirm the district court’s analysis of each Segment in their entirety and hold that it correctly identified and analyzed each segment under PPL. B. Weight of “Actual Use” Evidence Montana cites numerous instances of what it contends is evidence of “actual use” of the rivers that entitle Montana to a finding of navigability. Though Montana agrees that not just “any use” of a river qualifies, the claim that its “actual use” evidence by itself establishes navigability in this case is misguided. Montana does not cite any binding case indicating that actual use evidence is per se dispositive. One Oregon Court of Appeals case held that The Daniel Ball test is “disjunctive” and thus “either” actual use for travel and trade or susceptibility is sufficient. Nw. Steelheaders Ass’n, Inc. v. Simantel, 112 P.3d 383, 389 (Or. Ct. App. 2005). It is true that The Daniel Ball test states that rivers are navigable if “used, or are susceptible of being used . . . as highways for commerce.” 77 U.S. at 563 (emphasis added). But even the Oregon Court of Appeals hedged its bets, finding navigability after concluding the river at issue was both “susceptible to use for travel and trade at the time

STATE OF MONTANA V. TALEN MONTANA, LLC 21

of statehood” and “actually used for travel and trade” (for a certain portion of the river). Nw. Steelheaders, 112 P.3d at 393, 395. The Supreme Court has only stated that evidence of “extensive and continued” actual use may be the “most persuasive” for the navigability in fact analysis, rather than necessarily dispositive. Utah, 283 U.S. at 82 (emphasis added); see also PPL, 565 U.S. at 601 (noting that various uses may “bear upon susceptibility of commercial use” of rivers). To the contrary, it has held that “the true criterion” for determining river navigability is the river’s “capability of use by the public for purposes of transportation and commerce, rather than the extent and manner of that use.” Utah, 283 U.S. at 83; Montello, 87 U.S. at 441. To be sure, actual commercial use of a river that is more than occasional implies that the river has the capacity “to meet the needs of commerce.” Utah, 283 U.S. at 83; see also Rio Grande, 174 U.S. at 698 (evidence of occasional log drives insufficient). Thus, the “crucial question” is the rivers’ “susceptibility . . . to use as highways of commerce” at statehood rather than “the mere manner or extent of actual use.” Utah, 283 U.S. at 82 (emphasis added). Consider some examples from the evidence in this case. Montana cites the 1872 Thomas Roberts Survey to show “actual use” of the entirety of the Big Belt Mountains Segments. But that survey presented a mixed bag. Roberts noted there was “no business whatever done” on the river, he was not carrying any commercial cargo with him, and he had to portage around the Great Falls. He concluded that some steamboats with relatively shallow drafts could navigate the river, but also that any navigation required “great caution.” Roberts’s survey indicates that he made it through that portion of the river, but it cannot conclusively 22 STATE OF MONTANA V. TALEN MONTANA, LLC

establish—to the exclusion of conflicting evidence—that the Big Belt Mountains Segment could “meet the needs of commerce.” Utah, 283 U.S. at 83. Montana’s reliance on the journey of the Rose of Helena into the Great Falls area is likewise misplaced. The Rose made that trip once. Its captain, Nicholas Hilger, acknowledged it was a “dangerous undertaking,” while his son bluntly stated that “any man who has any judgment at all would have known better than to run those rapids.” 4 Other examples demonstrate the problem with Montana’s interpretation of actual use evidence. In 1891, Thomas Symons arguably “used” the entire Clark Fork River by surveying it, but he concluded it was “utterly unnavigable.” The record contains evidence of failed log drives near the Thompson Falls and the Eddy Segment and near the Great Falls, where someone either drowned or where they lost the logs over the falls. Such “uses” that confirm non-navigability or that end in tragedy are insufficient as a matter of law or common sense to deem a river segment navigable for title purposes. Evidence that rivers were actually used for trade and travel at statehood remains the “most persuasive” evidence of navigability, but does not necessarily tell the full story. The evidence must satisfy the “true criterion” of navigability, which is the river’s capacity to realistically meet and sustain the needs of “useful” commerce as the

4 The Rose retired to sporadically taking passengers through a six-to seven-mile stretch of the river to a picnic area near the Hilger ranch. Those trips occurred after the Army Corps of Engineers built some improvements to the river. Montana’s claim that it is entitled to this small portion of the river again misapplies both the segment-by- segment analysis and the navigability test.

STATE OF MONTANA V. TALEN MONTANA, LLC 23

Supreme Court has described it. See PPL, 565 U.S. at 600; Rio Grande, 174 U.S. at 698; Utah, 283 U.S. at 81–83, 86; Montello, 87 U.S. at 441, 443. We reject Montana’s arguments to the contrary. C. The District Court’s Findings of Non-Navigability The remaining bulk of Montana’s arguments wrangle with the district court’s factual findings and conclusions that the four Segments were not navigable. Montana ties these arguments to expert testimony the district court found unreliable, and to the mistaken premises that de novo review applies, that the analysis should only focus on the narrower Disputed Reaches, and that its evidence of “actual use” ends the inquiry. Nothing in the record leaves us with any impression— let alone a “definite and firm conviction”—that a reversible mistake occurred. Anderson, 470 U.S. at 573. The district court considered the extensive evidence and testimony before it and did not clearly err in its ultimate findings. For example, as to the Big Belt Mountains Segment, the district court considered such evidence as the 1872 Roberts survey, the Rose of Helena’s trip to Great Falls, conflicting reports and interpretations of Lewis and Clark’s journey, and competing expert testimony over the rivers’ depth and what kinds of watercraft, if any, could navigate the shallows and the “very hazardous” Beartooth Rapids. 5 In the Big Falls to Belt Creek Segment, Montana’s own expert testified that even smaller boats would “get stuck” and move around part of the Segment on land, directly undermining its position.

5 Contrary to Montana’s arguments, the district court did consider the ability of smaller watercrafts to navigate the challenging parts of the Segments, and ultimately concluded that Montana’s evidence was insufficient to demonstrate navigability for title. 24 STATE OF MONTANA V. TALEN MONTANA, LLC

In the Eddy Segment, the log drive evidence Montana relies on was inconclusive and largely occurred miles away from the Eddy Segment. And finally, the record supports the finding that the Headwaters/West Yellowstone Basin Segment was too sinuous and shallow to be a commercially viable river route. We find no clear error in the district court’s findings. At trial, the district court considered the evidence and simply credited Talen’s and NorthWestern’s experts over Montana’s and found Montana’s evidence insufficient to claim title. Disagreements with how the district court weighed the evidence or expert credibility are not grounds for reversal. Anderson, 470 U.S. at 573–74; Beech Aircraft, 51 F.3d at 838.

D. PPL’s Mandate and the “17-Mile Great Falls

Reach” Now we come to Talen’s and NorthWestern’s cross- appeal. They do not quibble with the district court’s finding that the Sun River to Black Eagle Falls Segment was navigable. Rather, they argue the district court should not have considered the Segment at all because it is part of the “17-mile Great Falls reach” that PPL held is not navigable. Montana disagrees, asserting that PPL’s navigability holding is limited and that the Segment lies beyond the “head of the first waterfall” that PPL described. The mandate rule states that when a higher court decides an issue and remands the case, that issue is “finally settled.” In re Sanford Fork & Tool Co., 160 U.S. 247, 255 (1895). The lower court is “bound by the decree as the law of the case” and cannot “vary [the decision], or examine it for any other purpose than execution.” Id. The lower courts on remand “must implement both the letter and the

STATE OF MONTANA V. TALEN MONTANA, LLC 25

spirit of the mandate, taking into account the [higher court’s] opinion and the circumstances it embraces.” Vizcaino v. U.S. Dist. Ct. for W. Dist. Wash., 173 F.3d 713, 719 (9th Cir. 1999). In this circuit, the mandate rule is jurisdictional—a mandate divests a lower court of jurisdiction to revisit the issue. See United States v. Thrasher, 483 F.3d 977, 982 (9th Cir. 2007). The mandate rule, however, only forecloses further consideration of what the “higher court decided, not . . . what it did not decide.” Kellington, 217 F.3d at 1093. Courts are still “free as to ‘anything not foreclosed by the mandate.’” Id. at 1092–93 (quoting Herrington v. Cnty. of Sonoma, 12 F.3d 901, 904 (9th Cir. 1993)). A district court’s decision “should not be reversed for failing to follow a mandate if its decision is within the scope of the remand.” Lindy Pen Co. v. Bic Pen Corp., 982 F.2d 1400, 1404 (9th Cir. 1993), overruled in part on other grounds by SunEarth, Inc. v. Sun Earth Solar Power Co., Ltd., 839 F.3d 1179, 1180–81 (9th Cir. 2016) (en banc) (per curiam). Our “ultimate task is to distinguish matters that have been decided . . . from matters that have not.” Kellington, 217 F.3d at 1093. Relevant here, the Supreme Court stated “. . . this Court now concludes . . . that the 17–mile Great Falls reach, at least from the head of the first waterfall to the foot of the last, is not navigable for purposes of riverbed title under the equal-footing doctrine.” PPL, 565 U.S. at 599 (emphasis added). The reach began “somewhat above the head of Black Eagle Falls,” and contains five waterfalls, around which Lewis and Clark portaged while exploring the territory in 1805. Id. at 583–84, 597. 26 STATE OF MONTANA V. TALEN MONTANA, LLC

A close reading of PPL reveals two clear mandates. First, the Supreme Court ruled that the portions of the river between the five waterfalls “from the head of the first . . . to the foot of the last” in the Great Falls area are not navigable. PPL, 565 U.S. at 599. Second, courts must apply the segment-by-segment approach, consider any need for and effect of land portages, and determine navigability based on a river’s “natural and ordinary condition” at statehood. See id. at 592, 593, 601. The Supreme Court reversed the state courts’ rulings due to “an infirm legal understanding of [the] Court’s rules of navigability for title under the equal-footing doctrine.” Id. at 604. The Court remanded the case “for further proceedings not inconsistent with this opinion.” Id. at 605. We conclude that the district court properly followed PPL’s mandate. The Supreme Court’s express limitation on its navigability ruling left the area upstream of the Black Eagle Falls open to further analysis. The Court could have held that the entire “17-mile Great Falls reach” was not navigable, but it did not. It instead chose to limit its holding to “at least from the head of the first waterfall to the foot of the last.” PPL, 565 U.S. at 599. The district court’s ruling is fully consistent with the “spirit” of PPL’s mandate for analyzing navigability as well. Vizcaino, 173 F.3d at 719. The Supreme Court reiterated the importance of precisely determining “the exact point at which navigability may be deemed to end.” PPL, 565 U.S. at 594 (quoting Utah, 283 U.S. at 90). The district court correctly recognized that the expert evidence “clearly raise[d] some issues of fact regarding the navigability of certain sections contained within the broadest reading of the 17-mile reach.” Defining and analyzing the “exact point” at which navigability starts and

STATE OF MONTANA V. TALEN MONTANA, LLC 27

ends beyond the non-navigable 8.2-mile stretch of waterfalls was thus proper and necessary to carry out the Supreme Court’s mandate. We conclude that the district court’s review and ruling of navigability of the Sun River to Black Eagle Falls Segment was consistent with the mandate from PPL.

IV. Conclusion

The law and the record of this case provide no support for any party’s arguments for reversal. We affirm the district court’s judgment in Phase I of this case. We remand to the district court to proceed with Phase II and to determine damages as to the Sun River to Black Eagle Falls Segment. The parties shall bear their own costs on appeal. AFFIRMED and REMANDED.