OPINION

Who would imagine that more than 150 years after the Treaty of Olympia (the “Treaty”) was signed between the United States and the Quileute and Quinault tribes, we would be asked to determine whether the “right of taking fish” includes whales and seals? Although scientists tell us sea mammals are not fish, these appeals ask us to go back to the 1855 treaty negotiation and signing and place ourselves in the ■ shoes of two signatory tribes—the Quileute Indian Tribe (the “Quileute”) and the Quinault Indian Nation (the “Quinault”)—to determine what they intended the Treaty to cover. In light of the evidence presented during the 23-day trial, the district court did not clearly err in its finding that the Quileute and-.Qui-nault understood that the Treaty’s preservation of the “right of taking fish” , includes whales and seals. The court’s extensive factual findings supported its ultimate conclusion that “ ‘fish’ as used in the Treaty of Olympia encompasses sea mammals and that evidence of customary harvest of whales and seals at and before treaty time may be the basis for the determination of a tribe’s [usual and accustomed fishing grounds].” We affirm the court’s judgment on that score. However, we reverse the court’s delineation of the fishing boundaries because the. lines drawn far exceed the court’s underlying factual findings.

Background

This appeal is one of many stemming from the long-running litigation over fishing rights in Western Washington. As we have noted, this litigation has a “lengthy background.” Tulalip Tribes v. Suquamish Indian Tribe, 794 F.3d 1129, 1131 (9th Cir. 2015). The story began in the mid-185.0s, when Governor Isaac Stevens approached the tribes of Western Washington with a proposal that the tribes cede most of their land to the United States but without giving up certain vital rights. His endeavor was successful: from December 1854 to January 1856, the United States entered into a series of similarly-worded treaties with the Washington tribes. Crucial to this appeal, the tribes preserved their right tp “tak[e] fish” at all “usual and accustomed grounds and stations.” That right has engendered a number of disputes between and among tribes about where each tribe can and cannot fish.

Here we address the Treaty of Olympia, which the Quileute and Quinault (as well as the Hoh Indian Tribe) sighed in July 1855. As with the other Stevens Treaties, the Treaty protects the tribes’ “right of taking fish at all usual and accustomed grounds and stations” (“U & A”). Treaty of Olympia, art. III, July 1, 1855-Jan. 25, 1856, 12 Stat. 971 , 972. In 1974, Judge Boldt of the Western District of Washington established standards and procedures for determining a tribe’s U & A and made U & A determinations for several tribes. United States v. Washington, 384 F.Supp. 312 (W.D. Wash. 1974) (Decision I), aff'd, 520 F.2d 676 (9th Cir. 1975).

This case is one in the ongoing saga arising from Judge Boldt’s original decision but presents a slight twist on the usual facts. Rather than asking whether the Quileute and Quinault have presented enough evidence to establish U & A in a particular location, the central issue here is whether evidence of hunting whales and seals can establish where the Quileute and Quinault were “taking fish” at and before treaty time.

Litigation on this issue began in 2009, when the Makah Indian Tribe (the “Ma-kah”) followed procedures to invoke the district court’s continuing jurisdiction to determine “the location of any of a tribe’s usual and accustomed fishing grounds not specifically determined” in Decision I. The Makah asked the district court to adjudicate the western boundary of the Qui-leute’s U & A and the Quinault’s U & A in the Pacific Ocean. The court held a 23-day trial—exceeding the length of Judge Boldt’s original trial leading to Decision I—and issued extensive findings.

Employing the Indian canon of construction, the court considered the Quileute and Quinault’s understanding of the Treaty’s ambiguous use of the word “fish” and found that, based on the historical and linguistic evidence, the tribes intended the term “fish” to encompass whales and seals. The court then looked at the evidence of pre-treaty Quileute and Quinault whaling and sealing and set the Quileute’s U & A boundary at 40 miles offshore and the Quinault’s U & A boundary at 30 miles offshore. Both the Makah and the State of Washington appeal.

Analysis

I. Evidence of Whaling and Sealing Is Appropriate to Establish U & A Under the Treaty of Olympia

A. Makah Is Not Law of the Case

The crux of this appeal is whether the term “fish” in the Treaty includes whales and seals. The Makah seeks to short-circuit the inquiry by reference to United States v. Washington (Makah), 730 F.2d 1314 (9th Cir. 1984). In the Makah’s view, we need not do much analytical heavy-lifting here because we already ruled in Makah that evidence of whaling and sealing cannot establish U & A. That reading of the case obscures what was actually decided and ignores a linchpin issue—in Makah we considered the Makah’s Treaty of Neah Bay, not the Treaty of Olympia.

The two treaties have an important textual difference: unlike the Treaty of Olympia, the Treaty of Neah Bay secures “[t]he right of taking fish and of whaling or sealing at usual and accustomed grounds and stations.” Treaty of Neah Bay, art. IV, Jan. 31, 1855, U.S.-Makah, 12 Stat. 939 , 940 (emphasis added). In addressing the Treaty of Neah Bay, we concluded that the Makah did not establish that its U & A extends 100 miles from the shore out to sea. Makah, 730 F.2d at 1318 . Given the express protection of the right to whale and seal, we had no need in Makah to separate out fishing from whaling and sealing or to address the significance of different types of evidence. It should be obvious that Makah is neither controlling nor informative because the question whether the Treaty of Olympia’s “right of taking fish” includes whales and seals was not “decided explicitly or by necessary implication.” United States v. Lummi Indian Tribe, 235 F.3d 443, 452 (9th Cir. 2000). Just as obviously, we cannot' simply transport analysis of the Treaty of Neah Bay-to the Treaty of Olympia because the member tribes’ intent is important to, if not dispositive of, the meaning of- particular provisions. See Choctaw Nation of Indians v. United States, 318 U.S. 423, 432 , 63 S.Ct. 672 , 87 L.Ed. 877 (1943) (holding that treaties involving Indian tribes “are to be construed, so far as possible, in the sense in which the Indians understood them ....”).

In Makah we described ..the question presented as “what ... .we find to be the Makahs’ usual and accustomed fishing areas.” 730 F.2d at 1316 . Consistent with that narrow framing of the issue on appeal, in discussing whether the Makah had presented sufficient évidence to establish its U & A out to 100 miles from shore, we explained:

Id. at 1318 .

The first paragraph hones in on the absence in the Makah’s evidence of regular fishing at 100 miles from shore. Although members of the Makah “were capable of traveling to 100 miles from shore” and “[t]hey may home canoed that far for whale and seal or simply to explore,” at the turn of the century ⅛ was “not clear how frequently” they fished at that distance. In contrast, we noted that “[ajbout 1900, they fished regularly at areas about 40 miles out, and. probably did so in the 1850’s.” Based on those facts and inferences, we held that the Makah’s U & A did not extend 100 miles into the ocean.

The concluding paragraph builds on that analysis, citing to Dr.- Lane’s suggestion that “the Makahs would travel [100 miles from shore] only when the catch was insufficient closer to shore.” Because “[t]he earliest evidence of insufficient catch” came “some 50 years after the treaty,” there was no basis to say that the Makah often traveled to the 100-mile mark at or before treaty time. The disparity between the Makah’s evidence with respect to 40 miles versus 100 miles drove our conclusion that the Makah did not “usually and customarily fish[ ] 100 miles from shore in 1855.”

This is not the first time that we have characterized Makah as turning on the extent of the evidence, presented. In an appeal involving the Tulalip Tribes, we noted that the “[e]vidence of frequent fishing- in the disputed areas is stronger ... than in the Makah case.” United States v. Lummi Indian Tribe, 841 F.2d 317, 320 (9th Cir. 1988). While the Makah’s evidence provided “no basis for an inference that [the Makah] customarily fished as far as 100 miles from shore at treaty time,” the Tulalip Tribes’ evidence “readily supported] an inference that the Tulalips frequently fished the disputed areas.” Id. This later case reinforces that Makah did not explicitly or implicitly decide the question of what role whaling and sealing evidence plays in a U & A determination, let alone address the Treaty of Olympia.

' B. The Treaty of Olympia Reserves the Quileute and Quinault’s Right to Take Whales and Seals

Having put the Makah case in context, wé turn to the interpretation of the Treaty of Olympia. The pertinent provision reads:

Treaty of Olympia, supra, 12 Stat. at 972 (emphasis added). The parties dispute whether the term “fish”—and the corresponding right to “tak[e] fish”—embraces whales and seals.

1. Textual Ambiguity

The text of the 'Treaty of Olympia does not nail down whether the term “fish” was meant to include or exclude whales and seals. At the time of signing, “fish” had multiple connotations of varying breadth. For example, Webster’s Dictionary simultaneously defined “fish” broadly as “[a]n animal that lives in water” (which would include whales and probably seals) and narrowly as a “name for a class of animals subsisting in water” that “breathe by means of gills, swim by the aid of fins,.and are oviparous” (which would exclude whales and seals). Webster’s American Dictionary of the English Language (1828). Other sources also acknowledged the popular understanding that the word “fish” could cover sea mammals; for example, the Supreme Court wrote that “For all the purposes' of common life, the whale is called a fish, though natural history-tells us that he* belongs to another order of animals.” In re Fossat, 69 U.S. 649 , 2 Wall. 649 , 17 L.Ed. 739 (1864).

The context in which the term “fish” is used does nothing to resolve the ambiguity. Although the Treaty preserves the “right of 'taking fish,” the action of “taking” is far-reaching and offers no meaningful constraint. Tribes may “tak[e]” whales and seals just as they may “tak[e]” fish. The shellfish proviso—-which prohibits the tribes from “tak[ing] shellfish from any beds staked or cultivated by citizens”—is similarly inconclusive, though it tends to point to a broader definition of fish. See United States v. Washington (Shellfish), 157 F.3d 630, 643 (9th Cir. 1998). We are left uncertain as to whether the Treaty employs the narrow or broad definition.

Nevertheless, the parties’ decision to employ capacious language, and particularly the expansive word “fish,” provides an indication of the provision’s comprehended scope. As we have recognized, if “the Treaty parties intended to limit the harvestable species, the parties would not have chosen the,word ‘fish’” because that word has “perhaps the widest sweep of any word the drafters could have chosen.” Id. (citation omitted). Notably, Judge Boldt’s original determination of the Quileute’s U & A relied on evidence of harvesting sea mammals. See Decision I, 384 F.Supp. at 372 (noting that “[a]long the adjacent Pacific Coast Quileutes caught ... seal, sea lion, porpoise and whale”).

The Makah does not advance a competing interpretation of the actual words of the Treaty of Olympia. Instead, it jumps to language in its own Treaty of Neah Bay, which explicitly references the right of “whaling [and] sealing” in addition to the right of “taking físh.” The Makah contends that, to avoid the problem of surplusage, the right of “taking fish” must be construed so as to exclude “whaling [and] sealing.” That argument is hard to swallow because we are not even talking about the same treaty.

As the district court observed, the Treaty of Neah Bay is of limited import because it “w[as] negotiated by different individuals and in [a] different context[ ].” Indeed, the “argument that similar language in two Treaties involving different parties has precisely the same meaning reveals a fundamental misunderstanding of basic, principles of treaty construction.” Minnesota v. Mille Lacs Band of Chippewa Indians, 526 U.S. 172, 202 , 119 S.Ct. 1187 , 143 L.Ed.2d 270 (1999). Rather than comparing and contrasting language and rights across treaties, courts “must interpret a treaty right in light of the particular tribe’s understanding of that right at the time the treaty was made.” United States v. Smiskin, 487 F.3d 1260, 1267 (9th Cir. 2007).

2. Indian Canon of Construction

Recognizing the ambiguity in the Treaty and underscoring that the Treaty of Neah Bay does not control interpretation of the Treaty of Olympia brings us to the Indian canon of construction. As a general rule, treaties “are to be-construed, so far as possible, in the sense in which the Indians understood them,” Choctaw Nation, 318 U.S. at 432 , 63 S.Ct. 672 , and “ambiguous provisions [should be] interpreted to their benefit,” Cty. of Oneida v. Oneida Indian Nation, 470 U.S. 226, 247 , 105 S.Ct. 1245 , 84 L.Ed.2d 169 (1985). That rule applies to “[t]reaty language reserving hunting, fishing, and gathering rights.” Cohen’s Handbook of Federal Indian Law § 18.02, at 1157 (Nell Jessup Newton ed., 2012). The Makah,- however, seeks to cut off the Quileute and Quinault’s argument from the get-go, asserting that the Indian canon does not apply here because “expanding] [the Quileute’s and Quinault’s] traditional fishing grounds adversely affects Makah.” The Makah’s contraction of the Indian canon is unwarranted.

Implicit in the Indian canon is the recognition that this principle inures to the benefit of the tribes that are parties to the treaty. As the Supreme Court has explained, the ultimate question is “how the [Indian] signatories to the Treaty understood the agreement because we interpret Indian treaties to give effect to .the terms as the Indians themselves would have understood them.” Mille Lacs, 526 U.S. at 196 , 119 S.Ct. 1187 (emphasis added). The canon is “rooted in the unique trust relationship” between the United States and the sovereign tribes, who stood in an unequal bargaining position. Cty. of Oneida, 470 U.S. at 247 , 105 S.Ct. 1245 ; Jones v. Meehan, 175 U.S. 1, 11 , 20 S.Ct. 1 , 44 L.Ed. 49 (1899). As a non-signatory party, the Makah cannot usurp application of the Indian canon with respect to the Treaty of Olympia. Such an incursion would undermine tribal sovereignty and the signatory tribes’ government-to-government relations. See Tavares v. Whitehouse, 851 F.3d 863, 877 (9th Cir. 2017); Cohen’s, supra, § 2.02, at 117.

The Makah reads our precedent too broadly to advocate for its seemingly limitless rule that the Indian canon is inapplicable whenever another tribe would be disadvantaged. Not surprisingly, the Ma-kah cites authority involving tribes claiming contradictory rights under the same statute or treaty; in those circumstances, the Indian canon is indeterminate because the government owes the same legal obligations to all interested tribes and “cannot favor one tribe over another.” Rancheria v. Jewell, 776 F.3d 706, 713 (9th Cir. 2015); Confederated Tribes of Chehalis Indian Reservation v. Washington, 96 F.3d 334, 340 (9th Cir. 1996).

Here, by contrast, we are faced with an interpretive choice that would favor the signatory tribes on the one hand and the United States on the other. See Rancheria, 776 F.3d at 713 . That conceptualization of the Indian canon also fits with Judge Boldt’s recognition that a tribe may establish U & A in an area “whether or not other tribes then also fished in the same waters.” Decision I, 384 F.Supp. at 332 . To the extent the Indian canon plays a part in understanding the Treaty, it is appropriate to invoke it here. We also note that we would reach the same .conclusion without a beneficial preference, as the evidence alone supports a broad interpretation of the Treaty language.

3. Intent of Quileute and Quinault

To ascertain the tribes’ understanding, courts “may look beyond the written words to the history of the treaty, the negotiations, and the practical construction adopted by the parties.” Water Splash, Inc. v. Menon, — U.S. —, 137 S.Ct. 1504, 1511 , 197 L.Ed.2d 826 (2017) (quoting Volkswagenwerk Aktiengesellschaft v. Schlunk, 486 U.S. 694, 700 , 108 S.Ct. 2104 , 100 L.Ed.2d 722 (1988)). After a 23-day bench trial, followed by 83 pages of Findings of Fact and Conclusions of Law, the district court undertook this task in a thoughtful and comprehensive manner.

Central to our review is the district court’s ultimate determination “that the Quinault and Quileute’s usual and accustomed fishing locations encompass those grounds and stations where they customarily harvested marine mammals—including whales and fur seals—at and before treaty time.” This conclusion rested on the extensive factual findings of the treaty negotiators’ intent—including the finding that the Quileute and Quinault understood the term “fish” covered whales and seals— and the underlying findings of historical fact, which were not clearly erroneous: See Shellfish, 157 F.3d at 642 .

The general context and tenor of the negotiations is a helpful starting point. Governor Stevens was appointed to negotiate with the tribes to extinguish their claims to Washington land and allow for peaceful cohabitation of Indians and non-Indians. During negotiations, the Indians’ main concern was reserving their “freedom to move about to gather food at them usual and accustomed fishing places” because harvesting fish was necessary for survival. Stevens and the other treaty commissioners made assurances throughout the process that the Indians would be able to continue their fishing activities and nowhere indicated that the Indians’ existing activities would be restricted or impaired by the treaties.

Stevens’s first attempt to reach an agreement with the Quinault in February 1855 at Chehalis River failed for reasons unrelated to this dispute. But in July 1855, the Quileute and Quinault (as well as the Hoh Indian Tribe) entered into the Treaty of Olympia, which protects the tribes’ “right of taking fish.”

The minutes from the failed negotiations offer some insight into key negotiating points, as the draft treaty from Chehalis River formed the basis for the negotiations of the Treaty of Olympia. Like Indians in other Stevens Treaty negotiations, the Indians at Chehalis River sought to preserve their entire subsistence cycle and worried that they would not be able to feed themselves if they ceded too much land. The commissioners explained that the treaty would confine where the tribes would live but would “not call[ ] upon [them] to give up their old modes of living and places of seeking food.” Stevens informed the tribes that the treaty “secures [their] fish” and permits them to “take fish where [they] have always done so and in common with the whites.”

Multiple aspects of the Treaty of Olympia negotiations shed light on the Quileute and Quinault’s understanding of the scope of “fish.” Although minutes from the negotiations do not exist today, the district court relied on ethnology studies and expert reconstructions of what likely happened at the negotiations. Because the commissioners and tribes did not speak the same languages, they used a limited trade medium of communication called Chinook jargon for translation. Colonel Shaw, the treaty commission’s official interpreter, translated provisions and remarks from English to Chinook jargon, then Indian interpreters translated the Chinook jargon into the tribes’ native languages.

One linguistic clue provides powerful evidence that the Quileute and Quinault assigned a broad meaning to the use of “fish.” The district court found, based on linguist Professor Hoard’s testimony, that “[t]he negotiators most likely used the Chinook word ‘pish,’” which translates into English as “fish.” The court credited Professor Hoard’s explanation that the negotiators would have opted for a broad cover term because Chinook language had general terms referring to large groups (like “fish”) and specific terms referring to individual species (like “whales,” “seals,” and “salmon”) but no intermediate terms referring to taxonomies (like “finfish” and “sea mammals”).

The Quileute’s and Quinault’s corresponding words for “pish” have even wider sweep. Like Chinook jargon, the Quileute and Quinault languages have no intermediate terms for taxonomies. As Professor Hoard explained, the Quileute would likely have used “?aalita?,” which translates as “fish, food, salmon.” Similarly, the Quinault’s term “Kémken” is defined alternatively as “salmon,” “fish,” and “food.” Because the Quileute and Quinault traditionally harvested whales and seals for food at and before treaty time, these pieces of linguistic evidence strongly support the district court’s finding that the tribes “would have understood that the treaty reserved to them the right to take aquatic animals, including ... sea mammals, as they had customarily done.”

The Makah counters that the Chinook, Quileute, and Quinault languages had separate words for “fish,” “whales,” and “seals” as well as for “fishing,” “whaling,” and “sealing.” But the mere existence of different words does not preclude some overlap in meaning. Such reasoning is as faulty as concluding that “tennis” and “volleyball” are not “sports” because “tennis,” “volleyball,” and “sports” are different words. Nor does the Makah’s identification of practical and cultural differences in the real-world occupations of fishing, whaling, and sealing bridge that gap. Additionally, that the tribes had distinct terms available does not undermine what terms were áctually utilized and how the Quileute and Qui-nault would have translated them. Because the Makah does not dispute that “pish” was used during negotiations and that “pish” can .mean something as broad as “food” in the Quileute and Quinault languages, it has not shown that the district court’s findings were erroneous, let alone clearly erroneous.

The district- court made extensive findings regarding fishing and subsistence activities at the time of the treaty. For both the Quileute and the Quinault, “fishing constituted the principle economic and subsistence activity ... at and before treaty time.” As to the Quinaultj “whale, seal, otter, :deer, bear, elk, sea-gulls, ducks, geese,” and “a variety of shellfish”, were among the wide range of animals harvested. Among other witnesses, Dr. Ronald Olson, an ethnologist, described in detail offshore fishing, whaling, and fur sealing.As to the Quileute, Judge Boldt recognized the significant role of oceanic resources and found that before and at treaty time, the Quileute harvested diverse resources, specifically singling out seal, sea lion, porpoise, and whale, among others. Supporting the link between food and whales, the district court related testimony that “[t]he Indians did not want all fish or all whale but liked to get something of everything which they wanted to eat.” Multiple witnesses contributed to the detailed findings on Quileute offshore fishing, whaling, and fur sealing.

Evidence of post-treaty activities further supports the view that the Quileute and Quinault (and possibly even the commissioners) understood the Treaty to protect whaling and sealing. No party contests the district court’s finding that “[djuring the post-treaty period, the[] tribes continued to harvest whales and seals from the Pacific Ocean” with active encouragement from government agents. Although the government’s acquiescence does not definitively show that the parties believed the right was preserved by the Treaty, the district court rightly noted that this important fact tends to suggest that “both sides believed the right to harvest sea mammals to have been reserved to the tribes.”

During the Chehalis River negotiations, neither the tribes nor the commissioners used the term “fish” in a manner inconsistent with its inclusion of whales and seals. The district court identified only two times where the tribes mentioned sea mammals explicitly—in both instances, the Indians asked for beached whales. Stevens answered one request for beached whales by stating that the. tribes “should have the right to fish in common with the whites, and get roots and berries.” Stevens replied to the other request with: “[The tribes] of course were to fish etc. as usual. As to whales they were theirs, but wrecks belonged to the owners Neither statement is clear as to whether Stevens understood fish and whales to be synonymous or overlapping, but we do not read his statements as drawing an incompatible distinction between the two. The broader understanding finds further support in a book by James Swan, who attended the negotiations and later wrote that “[t]he Indians, however,.were not to-be restricted to the reservation, but were to be allowed to procure their food- as they had always done.” •

As a practical matter, interpreting “fish” to cover whales and seals also respects the reserved-rights doctrine, which recognizes that treaties reserving fishing rights on previously owned tribal lands do not ’ constitute “a grant of rights to the Indians, bufa grant of right from them—a reservation of those not granted.” United States v. Winans, 198 U.S. 371, 381 , 25 S.Ct. 662 , 49 L.Ed. 1089 (1905); Cohen’s, supra, § 18.02, at 1156-57. In other words, absent a clear written indication, courts are reluctant to conclude that a tribe has forfeited previously held rights “because the United States treaty drafters had the sophistication and experience to use express language for the abrogation of treaty rights.” Mille Lacs, 526 U.S. at 195 , 119 S.Ct. 1187 . That doctrine favors reading the “right of taking fish” to include the Quileute’s and Quinault’s established historical whaling and sealing, particularly because there are independent indications that “fish” was understood that expansively. See Shellfish, 157 F.3d at 644 (employing the reserved-rights doctrine to assist in understanding the scope of a treaty provision' that could otherwise be read to encompass the right at issue). That practical point further solidifies that the Qui-leute and Quinault understood the “taking fish” provision to cover whales and seals.

Based on the considerable evidence submitted throughout the lengthy trial, the district court’s finding that the Quileute and Quinault intended the Treaty’s “right of taking fish” to include whales and seals was neither illogical, implausible, nor contrary to the record. We conclude that the district court properly looked to the tribes’ evidence of taking whales and seals to establish the U & A for the Quileute and the Quinault and did not err in its interpretation of the Treaty of Olympia, We do not address or offer commentary on whether the same result would obtain for the “right of taking fish” in other Stevens Treaties.

II, The Quileute and Quinault Have Identified the “Grounds and Stations” Where They Engaged in Whaling and Sealing

The State of Washington raises a separate argument, not joined by the Ma-kah, namely whether the Treaty of Olympia’s “grounds and stations” language mandates that the Quileute and Quinault provide evidence of “specific location[s] that the[y] regularly and customarily hunted whales or seals.” (Emphasis added). This argument falls into the sea.

The State’s suggestion that the tribes must identify specific named- locations directly conflicts with Judge Boldt’s description of “grounds and stations.” Judge Boldt defined “stations” as “fixed location's such as the site of a fish weir or a fishing platform or some other narrowly limited area” and “grounds” as “larger areas which may contain numerous stations and other unspecified locations which ,.. could not then have been determined with specific precision and cannot now be so determined.” Decision I, 384 F.Supp. at 332 .

While “stations” concerns particular locations and landmarks, “grounds” is not so limited. By definition, “grounds” includes “unspecified locations which ... could not then have been determined,” vitiating the State’s assertion that the tribes must come forward with specific named locations. The State’s claim also runs headlong into the practical reality that documentation of Indian fishing in 1855 is scarce, and requiring extensive and precise proof would be “extremely burdensome and perhaps impossible,” especially deep in the ocean. Shellfish, 157 F.3d at 644 . The district court appropriately-examined the substantial evidence of ocean whaling and sealing proffered by the Quileute and Quinault to determine that their usual and accustomed “grounds and stations” respectively-extend 40 miles offshore and 30 miles offshore.

III. The Longitudinal Lines Dp Not Match the District Court’s Findings

Having made U & A determinations for the Quileute and Quinault, the district court endeavored to draw precise boundaries where the tribes could fish. The parties agreed as to the northern boundaries but “dispute how the parties believe the Western boundary for the Quileute and Quinault should be demarcated as the line proceeds south.” The court decided to use longitudinal lines because it had done so in a prior proceeding with respect to the Makah’s boundaries. The court started at the northernmost point of the Quileute’s U & A, drew a line 40 miles west, and used that longitudinal position as the western boundary for the entire area. The court did the same with 30 miles for the Quinault. The map below depicts the final result.

[[Image here]]

. The Makah takes issue with the court’s use of a straight vertical line because the coastline trends eastward as one moves south. The Makah calculates the eoast-to-longitude distance at the southernmost point as 56 miles for the Quileute and 41 miles for the Quinault. In other words, the Quileute’s and Quinault’s southernmost boundaries respectively extend 16 miles and 11 miles beyond the court’s finding of usual and accustomed fishing, and their total areas respectively sweep in an extra 413 square miles (16.9% of the total 2,450 square miles) and 387 square miles (17.4% of the total 2,228 miles). The result would be different, for example, had the boundary lines been drawn parallel to the coastline.

These significant disparities underscore the deficiencies in the court’s longitudinal boundaries. The language of the Treaty of Olympia and countless judicial opinions spell out that the proceedings are designed to evaluate where the tribes were engaged in usual and accustomed fishing in 1855. After the court made that determination here, it effectively nullified parts of that same determination by creating a boundary containing large swaths of ocean where the Quileute and Quinault did not present sufficient evidence to establish U & A. Of course, practical difficulties mean that courts need not achieve mathematical exactitude in fashioning the boundaries. Nevertheless, the error rate here is too high and sweeps in areas that extend beyond the court’s factual findings. In our view, there are other solutions that better approximate the court’s findings.

The court’s stated reason for invoking longitudinal lines was that the approach “is the status quo method of delineating U & A ocean boundaries by this Court” and “equity and fairness demand the same methodology for delineating the boundary at issue here.” Although longitudinal lines were used to mark the Makah’s western boundaries in a separate case, nothing in that case suggests that longitudinal lines are the required methodology. See United States v. Washington, 626 F.Supp. 1405, 1467 (W.D. Wash. 1985). Notably, the court drew longitudinal boundaries there “[o]n the basis of all evidence submitted and reasonable inferences drawn therefrom .... ” Id. In denying a motion for reconsideration of the vertical boundaries, the court stated that the lines appropriately reflected “with some certainty the extent of the area which the Court intends to encompass within its determination of a tribe’s treaty-secured fishing area.” United States v. Washington, No. 70-9213, Dkt. # 8763, Mem. Op. on Mot. for Recons., at 2 (W.D. Wash. Jan. 27,1983). As shown in the map below, the lines tracked the coastline (and thus the court’s findings) in a way that avoids the problem presented by this case.

[[Image here]]

A different approach is warranted here to account for the dissimilarities between the eases. Although the Quileute and Qui-nault assert that the longitudinal lines also are appropriate because they are supported by the evidence, the boundaries do not reflect the district court’s findings. The Quileute and Quinault cannot vastly expand their U & A determinations without accompanying findings by the district court. Nor is the evidentiary gap solved by the court’s general statement that “tribal fishermen did not only fish due west of their villages, but moved in all directions from the coastline.”

Accordingly, we reverse the district court’s order imposing longitudinal boundaries. Because the law does not dictate any particular approach or remedy that the court should institute, we leave it to the court on remand to draw boundaries that are fair and consistent with the court’s findings.

AFFIRMED IN PART, REVERSED IN PART, AND REMANDED.

Each party shall bear its own costs on appeal.

. Modern popular culture recognizes that whales are mammals, not fish. An amusing exchange between two of the characters on Seinfeld provides one illustration; George: I’m such a huge whale fan. These marine biologists were showing how they communicate with each other with these squeaks and squeals, what a fish! Jerry: It's a mammal, George: Whatever. Seinfeld: The Marine Biologist (NBC television broadcast Feb. 10, 1994). . We refer to' the Treaty of Ólympia as a "Stevens Treaty,” as it was one of the similarly-worded treaties entered into by Governor Stevens between December 1854 and January 1856, In February 1855, Stevens negotiated with the Quinault a draft that formed the basis for the treaty negotiations. On July 1, 1855, Stevens sent Colonel Michael Simmons in his stead to negotiate the Treaty, which Stevens signed on Januaty 25, 1856. . One difference was that Colonel Simmons, sent by Governor Stevens to negotiate the Treaty of Olympia in Stevens’s stead, "lacked the authority to tailor provisions in the way that [ ] Stevens was able to do when negotiating the Treaty of Neah Bay." . The value of the minutes is somewhat diminished because the Quileute was not offi-dally, represented at this council; the tribe did, however, send along members to watch. . Because no party challenges the adequacy of the submitted whaling and seáling evidence, there is no basis to overturn the district court’s 40- and 30-mile findings. Nor do we need to reach the Makah’s and the State’s separate contention that the evidence was insufficient to establish that the Quileute’s customary finfishing extended 20 miles offshore.