Before King, Willett, and Douglas, Circuit Judges. Dana M. Douglas, Circuit Judge: In July 2023, Texas, at the direction of Governor Greg Abbott, installed a floating barrier in the Rio Grande near Eagle Pass, Texas. The United States filed a civil enforcement action against Texas, alleging that installment of the barrier violated the Rivers and Harbors Appropriation Act of 1899 (“RHA”). The United States moved for a preliminary injunction, which the district court granted, ordering the defendants to cease work on the barrier and to move it to the Texas riverbank. Texas timely appealed. This court entered an administrative stay. Finding that the district court did not abuse its discretion, we DISSOLVE the administrative stay and AFFIRM. I. BACKGROUND In early June 2023, Governor Abbott announced Texas’s intent to deploy “marine floating barriers” to “mak[e] it more difficult to cross the Rio Grande and reach the Texas side of the southern border.”[1] On July 10, 2023, without authorization from the federal government, Texas began installing the floating barrier. The floating barrier was deployed near Eagle Pass, Texas, with additional plans by Texas to deploy similar barriers at three different locations in the area. The floating barrier is roughly 1,000 feet long, made up of large four-foot orange buoys fastened together with heavy metal cables and anchored in place with concrete blocks placed systematically on the floor of the Rio Grande. The buoys are surrounded by 68 anchors weighing about 3,000 pounds each and 75 anchors weighing about 1,000 pounds each. Attached to about 500 feet of the floating barrier is a stainless-steel mesh “anti-dive net” extending two feet into the water. The following photographs are illustrative: On July 24, 2023, the United States brought a civil enforcement action under Sections 12 and 17 of theRHA,33 U.S.C. §§ 406, 413, seeking to enjoin the building of structures in navigable waters that obstruct the navigable capacity of those waters and to require Governor Abbott to remove all structures and obstructions in the Rio Grande. Specifically, the United States alleged that Texas violated Section 10 of the RHA, 33 U.S.C. § 403, by erecting a structure in the Rio Grande without authorization from the United States Army Corps of Engineers (the “Corps”) and creating an obstruction to the navigable capacity of that waterway without affirmative Congressional authorization. On September 6, 2023, after holding a hearing on the motion for a preliminary injunction filed by the United States, the district court concluded that the United States showed a likelihood of success on the merits and that the equities favored the granting of a preliminary injunction. It ordered Texas to cease any work on the floating barrier and to reposition it on the Texas side of the riverbank by September 15, 2023. This court administratively stayed the district court’s order pending consideration by this panel. We have jurisdiction under 28 U.S.C. § 1292(a)(1) from an order granting a preliminary injunction. II. STANDARD OF REVIEW “The decision to grant or deny a preliminary injunction is discretionary with the district court.” Miss. Power & Light Co. v. United Gas Pipe Line Co., 760 F.2d 618, 621 (5th Cir. 1985). Accordingly, we review a district court’s grant of a preliminary injunction for an abuse of discretion, reviewing factual findings for clear error and legal conclusions de novo. Harrison v. Young, 48 F.4th 331, 339 (5th Cir. 2022). “A finding of fact is clearly erroneous ‘when although there is evidence to support it, the reviewing court on the entire evidence is left with the definite and firm conviction that a mistake has been committed.’” Clark v. Mobil Oil Corp., 693 F.2d 500, 501-02 (5th Cir. 1982) (quoting United States v. U.S. Gypsum Co., 333 U.S. 364, 395 (1948)). III. DISCUSSION A plaintiff seeking a preliminary injunction must establish that “he is likely to succeed on the merits, that he is likely to suffer irreparable harm in the absence of preliminary relief, that the balance of equities tips in his favor, and that an injunction is in the public interest.” Winter v. Nat. Res. Def. Council, Inc., 555 U.S. 7, 20 (2008). A “preliminary injunction is an extraordinary remedy which should not be granted unless the party seeking it has ‘clearly carried the burden of persuasion.’” Harrison, 48 F.4th at 342 (quoting PCI Transp., Inc. v. Fort Worth & W.R.R., 418 F.3d 535, 545 (5th Cir. 2005)). The district court determined that the United States had carried its burden of persuasion as to all four requirements. A. Likelihood of Success on the Merits Section 10 of the RHA provides in pertinent part: The creation of any obstruction not affirmatively authorized by Congress, to the navigable capacity of any of the waters of the United States is prohibited; and it shall not be lawful to build or commence the building of any wharf, pier, dolphin, boom, weir, breakwater, bulkhead, jetty, or other structures in any . . . navigable river . . . of the United States. . . except on plans recommended by the Chief of Engineers and authorized by the Secretary of the Army…. 33 U.S.C. § 403. The first clause prohibits the construction of any obstruction in navigable waters without the consent of Congress. 33 U.S.C. § 403, cl. 1. The second clause prohibits the construction of specified and other structures in those navigable waters absent permission from the Corps. 33 U.S.C. § 403, cl. 2. The district court concluded that the United States showed a likelihood of success on the merits as to both clauses. For the United States to succeed on the first clause, the district court considered whether the segment of the Rio Grande at issue is a “navigable” waterway and whether the floating barrier is an “obstruction” to the navigable capacity of that waterway. For the United States to succeed on the second clause, the district court further considered whether the floating barrier is an “other structure” requiring a permit from the Corps before construction. We address each in turn. We then address the district court’s findings on Texas’s self-defense argument. i. Navigability To succeed under either clause of Section 10, the waterway must be navigable. Navigability is a factual question reviewed for clear error. United States v. Appalachian Elec. Power Co., 311 U.S. 377, 405 (1940). The Code of Federal Regulations defines navigable waters as: [T]hose waters that are subject to the ebb and flow of the tide and/or are presently used, or have been used in the past, or may be susceptible for use to transport interstate or foreign commerce. A determination of navigability, once made, applies laterally over the entire surface of the waterbody, and is not extinguished by later actions or events which impede or destroy navigable capacity. 33 C.F.R. § 329.4. Prior to the passage of the RHA, the Supreme Court defined navigable waters as those “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.” The Daniel Ball, 77 U.S. 557, 563 (1871). Following the passage of the RHA, the Supreme Court broadened its definition of navigability. In Economy Light & Power Co. v. United States, 256 U.S. 113 (1921), it expanded the definition to include waters formerly used in interstate commerce but no longer capable of such use. Id. at 123-24. And in Appalachian Electric, it expanded the definition to include waters not presently used, but capable of future use with reasonable improvement. 311 U.S. at 408-09. Thus, the elements to find navigability, as stated succinctly by the Sixth Circuit, require that the waterway “must (1) be or have been (2) used or susceptible of use (3) in the customary modes of trade and travel on water (4) as a highway for interstate commerce.” Mia. Valley Conservancy Dist. v. Alexander, 692 F.2d 447, 450 (6th Cir. 1982). Texas focuses much of its argument on the current state of the at-issue segment of the Rio Grande. But the current condition of the Rio Grande is not dispositive, as the tests for navigability set forth by the Supreme Court permit a review of whether the Rio Grande was historically navigable—that is, whether it was historically used in or susceptible of use in commerce—or whether it may be navigable in the future with reasonable improvements. See United States v. Utah, 283 U.S. 64, 82 (1931) (stating that “[t]he extent of existing commerce is not the test” for navigability); see also Econ. Light, 256 U.S. at 117 (finding the Desplaines River navigable despite “no evidence of actual navigation within the memory of living men”). The district court made two alternative, independent findings that the at-issue segment of the Rio Grande is navigable: first, that it had been used or was susceptible of use in commerce in the past and second, that it was susceptible of future use in commerce with reasonable improvements. We begin with historical navigability. The district court concluded that the United States presented sufficient evidence to carry its burden at this stage. We agree. In its complaint, the United States attached several exhibits discussing the navigability of the at-issue segment of the Rio Grande. Included among those is a document issued on December 20, 2011, by the Corps titled “Navigable Waters of the United States in the Fort Worth, Albuquerque, and Tulsa Districts Within the State of Texas.” It includes the Rio Grande from “the Zapata-Webb county line upstream to the point of intersection of the Texas-New Mexico state line and Mexico” as “navigable waters of the United States” falling within the Corps’ jurisdiction.[2] The United States also attached an authenticated document titled “Navigability Determination, Rio Grande River, TX,” published in 1984, in which the United States Coast Guard indicated that the Rio Grande River “was listed among the navigable waters of the United States pursuant to treaties with Mexico and for Coast Guard regulatory purposes.” The Coast Guard published these determinations yearly from 1947 to 1976, when the practice of publishing lists of navigable waters ceased. The document further stated that “the designated stretch of the Rio Grande River remains a navigable waterway of the United States.”[3] In his declaration authenticating the document, Captain Brandy Parker of the U.S. Coast Guard indicated that “the 1984 navigability determination is still in effect.” The United States also relied on an extensive study performed by the Corps that concluded: “The Rio Grande River between River Mile 275.5 and 610.0, on the United States side from the centerline of the normal channel, is a navigable water of the United States.” This determination of navigability by the Corps is based on a report titled “Navigability Study, Rio Grande, Tributaries, and Lakes, Rio Grande Basin, River Mile 275.5 to 610.0 date March 1975″ (the “Navigability Study”). The Navigability Study, relied upon by the district court, reflects in detail on past commerce in the region.[4] The first practical navigation in the region occurred during the Mexican American War. “General Zachary Taylor importuned the quartermaster for light steamboats for patrol and military transportation purposes. The Corvette, Whitesville, Major Brown, and Colonel Cross arrived from Pittsburg in June, 1846. Other steamboats plying the Rio Grande at the time were the J. E. Roberts and the Brownsville.” In an extract from People and Plots on the Rio Grande by V. N. Lott and V. M. Fenwick (1957), the authors indicate the importance of steamboats in the transportation of supplies for Taylor’s army: “These played a most important role in the history of the conflict and for many years thereafter river navigation was vital to the life of the area between Ringgold and Brownsville, and to some extent as far up to Laredo.” Navigation was then extended further upstream: “In October, 1846, a successful attempt was made to ascend the Rio Grande in the United States steamer Major Brown, by order of General Patterson . . ..” Another account discusses navigating to the Big Bend reach, the farthest upstream area of which any account was found: The head of steam navigation in the 1850′s . . . was Roma. Goods were trans-shipped upriver to the forts by pack and wagon from there . . .. First of all, until steamboats were released from wartime troop duty, he supplied the river garrisons by keelboat. And then in 1850 he sent an expedition up the river with orders to navigate to the farthest possible place. He hoped to discover that shipping could utilize far more of the river’s length than it had so far done. A keelboat and a skiff, manned by sixteen men, ascended the river by channel to a point a thousand miles above the head of steam travel or about thirteen hundred miles above the mouth. It was an astonishing penetration for a river with so little water, and the expeditioneers came back, all safe, to report optimistically that if the channel were improved in certain passages, steam navigation would be entirely feasible all the way “up to Babbitt’s Falls.” Also included in the Navigability Study is an extract from the 1949 book Rio Grande, River of Destiny by Laura Gilpin that states: In 1861 the Rio Grande was navigable for two hundred miles, and Texas-grown cotton was brought to the mouth of the river by small boats for the transfer to ocean-going vessels in the Gulf of Mexico. Down the coast in small boats and across the land by wagon came cotton to be sold to Europe. Up the coast and back across the land went supplies for the Confederate Army, as these were imported from England, France, and Mexico. In an extract from Southwestern Historical Quarterly (1914) edited by Ephraim Douglass Adams, a letter titled “British Correspondence Concerning Texas” sent in 1843 states: All the rivers however, discharging themselves into the Gulf, vary greatly in their navigable facilities, according to the season, and I dare say, that in the winter and spring Months, the Rio Grande would be navigable for a great distance in light iron boats, such as are used in the upper Ganges and Indus. . . . Matamoros, Tampico, Alvarado, Tabasco are all accessible to Vessels of the draught I have indicated. The district court, as well as the Corps’ Navigability Study, also considered Article VII of the Treaty of Guadalupe Hidalgo between the United States and Mexico, in which the countries agreed that navigation on the Rio Grande “divided in the middle between the two Republics. . . shall be free and common to the vessels and citizens of both countries; and neither shall, without the consent of the other, construct any work that may impede or interrupt, in whole or in part, the exercise of this right.” 9 Stat. 928 (1848) (free navigability reinforced by the Gadsden Treaty, art. IV, 10 Stat. 1034 (1853)). Contemporaneous court cases, cited by the district court, also support a finding of historical navigability. These cases discuss ferry companies operating between Eagle Pass and Piedras Negras, transporting goods in commerce, such as cotton. United States v. Weil, 35 Ct. Cl. 42, 77 (1900) (“At Eagle Pass there were ferryboats in which cotton was crossed over. . ..”); Tugwell v. Eagle Pass Ferry Co., 9 S.W. 120 (Tex. 1888) (resolving dispute between ferry companies operating between Eagle Pass and Piedras Negras). Contrary to the dissent’s position, ferry usage is an appropriate consideration when evaluating historical navigability. See Appalachian Elec., 311 U.S. at 413 n.46 (considering ferry usage). Moreover, the dissent points to a Georgia district court decision—United States v. Crow, Pope & Land Enters., Inc., 340 F. Supp. 25, 35 (N.D. Ga. 1972)—to support its contention that ferry usage does not demonstrate commerce in the region. But Crow considered intrastate bank-to-bank ferry traffic occurring entirely within Georgia on the Chattahoochee River, not commerce across an international boundary. See id. at 29-30. The ferry cases cited by the United States specifically involve the transportation of goods in commerce, such as cotton, between the United States and Mexico—a feature that certainly makes the Rio Grande unique and undermines any contention that navigation must be along the river.[5] In addition to the agency determinations, historical evidence, treaties between the United States and Mexico, and court cases, the district court noted several acts of Congress, attached to the United States’ complaint, that emphasize the importance of navigability in the region. In one act to authorize the construction of a bridge over the Rio Grande between the cities of Eagle Pass, Texas, and Piedras Negras, Mexico, Congress states: [S]aid bridge shall not interfere with the free navigation of said river, and in case of any litigation arising from an obstruction or an alleged obstruction to the free navigation thereof, caused or alleged to be caused by said bridge, the case may be tried before the circuit or district court of the United States for the State of Texas having jurisdiction thereof. 23 Stat. 29 (1884). It further states: Congress reserves the right to withdraw the authority and power conferred by this act, in case the free navigation of said river shall at any time be substantially or materially obstructed by said bridge, or for any other reason, and to direct the removal or necessary modifications thereof at the cost and expense of the owners of said bridge. . .. Id.[6] A few years later, Congress authorized the Texas-Mexican Electric Light and Power Company to erect wires across the Rio Grande at Eagle Pass, Texas: Provided, That said wires shall not interfere with the free navigation of said river, and in case of any litigation arising from an obstruction or alleged obstruction to the free navigation thereof, caused or alleged to be caused by said wires, the case may be tried before the district court of the United States for the western district of Texas: And provided also, That Congress reserves the right to withdraw the power and authority conferred by this act incase the free navigation of the river shall at any time be substantially or materially obstructed by said wires, or for any other reason, and to direct the removal of said wires, or necessary modifications thereof, at the cost and expense of the owners of said wires . . . .. 26 Stat. 495 (1890).[7] Perhaps if all the United States identified at this stage were these acts of Congress, we might find clear error based on Oklahoma v. Texas, in which the Supreme Court, evaluating similar acts, determined that these provisions were “only precautionary” and “not intended as an affirmation of navigable capacity in that locality.”[8] 258 U.S. 574, 586 (1922). But the district court relied on additional evidence, outlined supra, of the same sort relied upon by the Supreme Court in Appalachian Electric: Use of a stream long abandoned by water commerce is difficult to prove by abundant evidence. Fourteen authenticated instances of use in a century and a half by explorers and trappers, coupled with general historical references to the river as a water route for the early fur traders and their supplies in pirogues and Durham or flat-bottomed craft similar to the keelboats of the New [River], sufficed upon that phase in the case of the DesPlaines. 311 U.S. at 416 (emphasis added). Likewise, here, there are numerous historical references regarding use in commerce throughout the 1800s. Additionally, the Supreme Court in Appalachian Electric stressed the broadness in which it finds navigability: “Nor is it necessary for navigability that the use should be continuous. The character of the region, its products and the difficulties or dangers of the navigation influence the regularity and extent of use. Small traffic compared to the available commerce of the region is sufficient.” Id. at 409 (emphasis added). “Even absence of use over long periods of years, because of changed conditions, the coming of the railroad or improved highways does not affect the navigability of rivers in the constitutional sense.” Id. at 409-10. “Nor is lack of commercial traffic a bar to a conclusion of navigability where personal or private use by boats demonstrates the availability of the stream for the simpler types of commercial navigation.” Id. at 416 (emphasis added). Finally, the district court’s finding of historical navigability on the at-issue portion of the Rio Grande is supported by the Supreme Court’s holding in United States v. Rio Grande Dam & Irrigation Co., 174 U.S. 690 (1899), relied upon by the district court and the Corps’ Navigability Study. There, the Supreme Court held that the Rio Grande River is “not navigable within the territory of New Mexico.” Id. at 695. However, in discussing the propriety of taking judicial notice of navigability within the territory of Texas, the Supreme Court expressly stated: “That the Rio Grande, speaking generally, is a navigable river, is clearly shown by the affidavits. It is also a matter of common knowledge, and therefore the courts may properly take judicial notice of that fact.” Id. at 698.[9] Accordingly, we find no clear error in the district court’s historical navigability finding. The dissent would demand a higher showing than what is required. To succeed under the RHA, the United States need not prove that the Rio Grande was, in fact, historically used for commerce. It need only show that it was historically susceptible of use for commerce. The evidence put forth by the United States overwhelmingly supports, at least, the conclusion that the at-issue segment of the Rio Grande was historically susceptible of use for commerce. On these facts, we are not left with a “definite and firm conviction” that the district court erred. Clark, 693 F.2d at 501-02 (quoting U.S. Gypsum Co., 333 U.S. at 395). Moreover, we expect the evidence to develop as this case proceeds to the merits. We reiterate the posture of the case before us. We are not reviewing a permanent, but a preliminary, injunction. The “extraordinary” nature of a preliminary injunction does not heighten Winter‘s “likely” standard. See Winter, 555 U.S. at 22. For now, this showing is sufficient.[10] ii. Obstruction Section 10′s first clause bars the “creation of any obstruction not affirmatively authorized by Congress, to the navigable capacity of any of the waters of the United States.” 33 U.S.C. § 403. After finding navigability, the district court found obstruction, concluding that “the floating barrier interferes with or diminishes the navigable capacity of the Rio Grande and creates a hazard.” Whether there is an obstruction is a question of fact reviewed for clear error. Rio Grande Dam, 174 U.S. at 709. Texas argues that to be an obstruction, an object must tend to destroy the navigable capacity of a waterway. To the contrary, the Supreme Court has defined an obstruction as tending to “interfere with or diminish[] the navigable capacity of a stream.” Id. It has also emphasized that its own cases define “obstruction” as used in Section 10 as “broad enough to include diminution of the navigable capacity” of the waterway at issue. United States v. Republic Steel Corp., 362 U.S. 482, 489 (1960).[11] Under this broad construction, the Supreme Court has previously found matter described as “fine particles” from an iron mill to be an obstruction under Section 10. Id. at 483. The Fifth Circuit has likewise acknowledged the Supreme Court’s broad interpretation of obstruction. Vieux Carre Prop. Owners, Residents & Assocs., Inc. v. Brown, 875 F.2d 453, 462-63 (5th Cir. 1989) (“The Supreme Court has encouraged a broad interpretation of a section 10 ‘obstruction’. . ..”). And it has construed the term flexibly, as the district court noted, without a size or positional limit. See United States v. Raven, 500 F.2d 728, 731 (5th Cir. 1974) (finding a sunken schooner to be an obstruction “[i]f floating particles can be an obstruction”); United Tex. Transmission Co. v. U.S. Army Corps of Eng’rs, 7 F.3d 436, 438 (5th Cir. 1993) (finding a pair of gas pipelines running under the bed of a bayou an obstruction). Here, the district court began its analysis by noting that Texas designed and deployed the floating barrier to literally obstruct lateral movement across the river.[12] It then looked to the credible testimony and evidence before it. The declaration of Mario Gomez, Acting Area Operations Manager for the Amistad Dam Field Office of the United States Section of the International Boundary and Water Commission (“IBWC”),[13] indicated that “[n]ormally, the Mexico or U.S. Section of the Commission can go into any location of the Rio Grande independently and do surveying and other engineering work that the Commission Sections carry out” but that the floating barrier is “an impediment to the Sections crossing independently in this part of the river,” including a planned survey by the Mexican Section of the Commission that was unable to proceed because of the obstruction. Likewise, the Chief of the U.S. Border Patrol (“USBP”), Jason D. Owens, declared that border patrol agents rescue individuals in distress in the Rio Grande, utilizing “small watercraft to quickly respond as the incidents unfold.” He noted that “[a]ny obstructions in the water could naturally impair the freedom of movement and potentially delay response times.” “From the beginning of fiscal year 2018 through July 23, 2023 there were 249 water-related rescues and 89 water-related deaths of individuals whose rescue or death occurred in or around the Rio Grande throughout the Eagle Pass Station AOR.”[14] Further, in the declaration of Joseph L. Shelnutt, the Regulatory Project Manager in the Compliance and Enforcement Branch for the Corps, he indicates that the “placement and tandem configuration of the buoys, which allows them to move somewhat independently even though they are connected, present a structural barrier to cross-river navigation and would force a vessel to maneuver around the structure to avoid collision or entanglement at this location.” The district court also noted that the floating barrier is not simply a string of buoys but is made even larger by the anchors placed four to six feet on either side of the barrier: “Photographs show these grey concrete anchors standing from the bed of the river, with no markings to identify them as hazards. These concrete obstacles present a serious risk to watercraft of any kind.” This is because the anchors are not easily seen by oncoming watercraft but are at a level that would cause damage to a vessel of any size that came upon them. Texas’s own declarants state that the portion of the Rio Grande at issue has many hazards, including “sand bars, shallow water, water with inconsistent depths, small islands, large rocks, man-made debris, natural debris such as logs and stumps, and sandy shoals” and that it is “very difficult and dangerous” even for airboats to operate. As the district court found, these conditions “make it even more imperative for anyone piloting down the river to have free reign [sic] of the entire width and a clear view of all obstacles.” This evidence, coupled with the Supreme Court’s command to interpret “obstruction” within Section 10 broadly, supports the district court’s finding that the floating barrier is an obstruction to the navigable capacity of the Rio Grande, and we find no clear error. Having made the requisite showing that Texas is likely in violation of the first clause of Section 10 because its obstruction was not “affirmatively authorized by Congress,” 33 U.S.C. § 403, the United States has shown that it is likely to succeed on the merits of its first claim. iii. Other Structure Section 10′s second clause prohibits “build[ing] or commenc[ing] the building of any wharf, pier, dolphin, boom, weir, breakwater, bulkhead, jetty, or other structures in any. . . navigable river. . . except on plans recommended by the Chief of Engineers and authorized by the Secretary of the Army.” 33 U.S.C. § 403. The Corps defines “structure” expansively including, “without limitation, any pier, boat dock, boat ramp, wharf, dolphin, weir, boom, breakwater, bulkhead, revetment, riprap, jetty, artificial island, artificial reef, permanent mooring structure, power transmission line, permanently moored floating vessel, piling, aid to navigation, or any other obstacle or obstruction.” 33 C.F.R. § 322.2. The district court found that the floating barrier was a “boom” or “other structure,” requiring a permit prior to construction, and thus that the United States was likely to prevail on its Section 10, clause 2 claim. We agree that it is an “other structure,” and do not consider whether it is a “boom.” As the United States argues, all the structures listed in Section 10 are built in water and tend to be obstacles or obstructions to navigation. In other words, these structures are all tangible objects that “interfere with or diminish” navigation by requiring vessels to move around them. See Rio Grande Dam, 174 U.S. at 709 (defining “obstruct”). The barrier fits within this broad definition because vessels must navigate around the barrier, and some may even be completely thwarted by its presence. Texas’s argument that to constitute an “other structure,” the floating barrier must be permanent is unconvincing. Even if the other enumerated structures were permanent, Texas has not shown that the floating barrier is not. The barrier has a “tamper resistant” design and “heavy concrete blocks placed systematically on the bed of the Rio Grande River to prevent movement.” And the barrier “is meant to withstand at least a 100-year flood” and “rise and fall with the elevation of the water while maintaining the same position on top of the river.” As the district court noted, “Texas’s own declarants attest that it would take ‘several weeks,’ heavy equipment, and $300,000 to remove the barrier.” Moreover, Texas has no intention of removing the barrier after a short time, as is made clear by the current litigation. Having also made the requisite showing that Texas is likely in violation of the second clause of Section 10 because the floating barrier is an “other structure” built without seeking Corps approval, the district court correctly determined that the United States has shown it is likely to succeed on the merits of both its RHA claims. iv. Invasion Texas argues that the Constitution gives it a right to defend itself from “invasion,” so it should be exempt from any RHA violations. U.S. Const. art. I, § 10, cl. 3 (“No state shall, without the Consent of Congress, . . . engage in War, unless actually invaded, or in such imminent Danger as will not admit of delay.”). We note that a “preliminary injunction may issue . . . despite the existence of a plausible defense, as long as the movant demonstrates a substantial likelihood of success.” Dall. Cowboys Cheerleaders, Inc. v. Scoreboard Posters, Inc., 600 F.2d 1184, 1188 (5th Cir. 1979). We have already affirmed the district court’s determination that the United States showed a substantial likelihood of success on the merits. However, “the district court must at least make clear that it has considered plausible defenses which are fully briefed and argued by [the] defendants.” Allied Mktg. Grp., Inc. v. CDL Mrkt., Inc., 878 F.2d 806, 815 (5th Cir. 1989). “While a preliminary injunction may be appropriate even in the face of potentially significant defenses, it is frequently desirable in such cases to expedite the trial on the merits.” Id. (citing 11 Charles Alan Wright & Arthur R. Miller, Federal Practice & Procedure § 2950, at 484 (1973 & Supp. 1988)); see also Charles A. Wright, Arthur R. Miller & Edward H. Cooper, Federal Practice & Procedure § 3291.1 (2d ed. 2011) (“Ordinarily the scope of appellate review under § 1292(a)(1) is confined to the issues necessary to determine the propriety of the interlocutory order itself. The curtailed nature of most preliminary injunction proceedings means that the broad issues of the action are not apt to be ripe for review. . ..”). Here, Texas’s self-defense argument does not preclude the issuance of a preliminary injunction. First, the district court adequately considered Texas’s arguments. Further, the district court, mindful of the sensitive nature of the parties’ interests in this case, sought to expeditiously seek a determination on the merits. Finally, the gravity of Texas’s argument— particularly the constitutional implications of a single state’s ability to declare it is invaded and select its own means of waging war—suggests it would be best considered on a fully developed record. “The purpose of a preliminary injunction is to prevent irreparable injury so as to preserve the court’s ability to render a meaningful decision on the merits.” Miss. Power & Light, 760 F.2d at 627 (citing Canal Auth. of State of Fla. v. Callaway, 489 F.2d 567, 576 (5th Cir. 1974)). The evidence before the district court did not show that the floating barrier had any meaningful impact on deterring any perceived “invasion,” so its removal is unlikely to cause Texas irreparable injury.[15] Accordingly, the preliminary injunction appropriately maintains the status quo while the district court expeditiously considers the case on its merits. B. Balance of Equities Since its issuance, Winter v. Natural Resources Defense Council, Inc., 555 U.S. 7 (2008), has been the guidepost for determining whether a plaintiff has made an adequate showing to warrant entry of a preliminary injunction. The United States has shown a likelihood of success on the merits, so we turn to whether it has shown a likelihood of irreparable harm absent preliminary relief, that the balance of equities tips in its favor, and that an injunction is in the public interest. Id. at 20. The district court, in the alternative to its holding that it need not consider these additional factors where the United States enforces a public-interest statute, expressly addressed each one.[16] Its balancing was not an abuse of discretion. First, the district court considered whether the United States had demonstrated “that irreparable injury is likely in the absence of an injunction.” Winter, 555 U.S. at 22. It pointed to the “tremendous strain” on the U.S.-Mexico relationship. It further found that the floating barrier “threatens the IBWC’s ability to implement the core provisions of the 1944 Treaty between the United States and Mexico, which is crucial to allocation of waters in the Rio Grande,” and that one meeting concerning water releases was already cancelled by the Mexican Section of the IBWC due to the presence of the floating barrier. The district court also found that the floating barrier posed a risk to human life, which is supported by Texas’s own statements noting the treachery of venturing across the Rio Grande.[17] The weight of the evidence before the district court supports its conclusion. The declaration of Jennifer T. Pena, Chief Legal Counsel for the United States Section of the IBWC, discussed the collaboration between the United States and Mexico. In a 1944 Treaty between the United States and Mexico called the “Utilization of Waters of the Colorado and Tijuana and of the Rio Grande,” the two countries agreed “how to allocate the waters of the bi-national rivers in the Treaty title, including the Rio Grande.” In a 1970 Treaty to Resolve Pending Boundary Differences and Maintain the Rio Grande and Colorado River as the International Boundary, both countries agreed “that the international boundary will be the middle of the channel occupied by the normal flow of the river.” Because river channels may shift over time, the countries agreed to take measures to stabilize and prevent shifts, including by “prohibit[ing] the construction of works in its territory which . . . may cause deflection or obstruction of the normal flow of the river or of its flood flows.” The floating barrier “has been the subject of a series of correspondence from the Mexican section of the IBWC” and is “interfering with the ability of the IBWC to fulfill its mission.” On July 14, 2023, Mexico’s Section of the IBWC “objected to the placement of the buoys and requested intervention of the United States Section to remove the buoys.” It further reported that, “as a result of the floating barriers in Eagle Pass, Mexico was cancelling a July 24, 2023, meeting concerning water releases to the United States from the Rio Conchos River in Mexico.” In a July 14, 2023, meeting between the Texas Commission on Environmental Quality and the IBWC to discuss delivery by Mexico to the United States of water from the Rio Grande, Mexico’s Section of the IBWC indicated that Texas’s unilateral actions “could affect cooperation between the two countries going forward.” Currently, the United States and Mexico are in the midst of trying to reach an agreement by December 2023 on a “new mechanism to improve the predictability and reliability of Rio Grande water delivery from Mexico to the United States” and discussions are “at a sensitive stage.” “By causing Mexico to rethink and limit its cooperation with the United States, Texas’s floating barrier interferes with IBWC’s ability to implement its core mission of implementation of the 1944 Water Treaty for deliveries of water on the Rio Grande.” “[I]f the proposed [structure] and appropriation of waters of the Rio Grande constitute a breach of treaty obligations or of international duty to Mexico, they also constitute an equal injury and wrong to the people of the United States.” Rio Grande Dam, 174 U.S. at 701. On July 24, 2023, Hillary Quam, the U.S.-Mexico Border Coordinator at the U.S. Department of State and acting Director of the Office of Mexican Affairs, declared that if the barrier “is not removed expeditiously, its presence will have an adverse impact on U.S. foreign policy,” and that Mexico “[o]n a number of different occasions beginning in late June, 2023″ has “protested to the United States the deployment of a floating barrier within the Rio Grande.” Mexico has protested the installation of the floating barrier, “asserting that it causes obstruction and deflection of the river as well as possible runoff into Mexican territory” “in contravention of the provisions of the 1970 Treaty.” In a supplemental declaration on August 15, Ms. Quam indicated as follows: The Government of Mexico has continued to protest to the United States the deployment of the barrier. On August 10, the Secretary of State held a meeting in Washington, D.C., with his counterpart from Mexico, Foreign Secretary Alicia Bárcena . . . The topic of the floating barrier was the first one raised by Foreign Secretary Bárcena. The Foreign Secretary indicated that Mexico is very concerned about the floating barrier and grateful for this lawsuit. The President of Mexico, Andrés Manuel López Obrador, in his daily press conference, has discussed the matter of the floating barrier at least six times since July 25, 2023. Crediting this evidence to find a likelihood of irreparable harm was not erroneous.[18] Turning to the balance of equities and the public interest, when the United States is a party, the third and fourth elements of the traditional preliminary injunction analysis merge. Cf. Nken v. Holder, 556 U.S. 418, 435 (2009). The district court found that “the barrier’s threat to human life, its impairment to free and safe navigation, and its contraindication to the balance of priorities Congress struck in the RHA outweigh Texas’s interest in implementing its buoy barrier in the Rio Grande River.” The district court, in finding that the public interest favored the United States, emphasized Supreme Court decisions. In Sanitary District of Chicago v. United States, 266 U.S. 405 (1925), the Supreme Court stated: “There is no question that this power [to remove obstructions to interstate and foreign commerce] is superior to that of the States to provide for the welfare or necessities of their inhabitants.” Id. at 426. More pointedly, in Arizona v. United States, 567 U.S. 387 (2012), the Supreme Court noted that while Arizona “may have understandable frustrations with the problems caused by illegal immigration,” as Texas has alleged, “the State may not pursue policies that undermine federal law.” Id. at 416. Further, the district court emphasized that the “balance of priorities Congress struck in the RHA” outweighed Texas’s asserted interests. Courts may look at the statute at issue for guidance in determining whether the issuance of a preliminary injunction would be in the public interest. See, e.g., Amoco Production Co. v. Village of Gambell, AK, 480 U.S. 531, 544 (1987) (finding the public interest promoted by the Alaska National Interest Lands Conservation Act was “to protect Alaskan subsistence resources from unnecessary destruction,” rather than preventing the actions the plaintiff sought to enjoin). Congress has spoken to the public interest through passage of the RHA, and the Supreme Court has emphasized the same: “We are dealing here with the sovereign powers of the Union, the Nation’s right that its waterways be utilized for the interests of the commerce of the whole country.” Appalachian Elec., 311 U.S. at 405.[19] The district court relied on all the evidence discussed herein to find that the balance of hardships tips in favor of the United States. It considered the threat to navigation and federal government operations on the Rio Grande,[20] as well as the potential threat to human life the floating barrier created. All of the district court’s findings of fact were well supported by the record, and its conclusion that the equities favor issuance of a preliminary injunction was not an abuse of discretion. IV. CONCLUSION The district court’s factual findings were not clearly erroneous. Its grant of a preliminary injunction was not an abuse of discretion, as the United States carried its burden. Accordingly, we DISSOLVE the administrative stay and AFFIRM. Texas’s motion to stay the district court’s preliminary injunction pending appeal is DENIED AS MOOT. Don R. Willett, Circuit Judge, dissenting: In response to millions of illegal border crossings in recent years, accompanied by millions of pounds of smuggled drugs, Texas Governor Greg Abbott deployed a floating barrier along a 1,000-foot stretch of the Rio Grande near Eagle Pass, a popular crossing spot. The bright orange chain of tethered buoys—a floating border wall anchored to the riverbed by concrete blocks—is Governor Abbott’s latest effort to quell the record influx of illegal crossings, something he has declared an “invasion.” Today, the majority opinion upholds the district court’s grant of a preliminary injunction requiring Texas to cease work on the barrier and shift it to the Texas riverbank. I disagree and would vacate the injunction entered by the district court, as the United States has failed to carry its burden on any of the four preliminary injunction factors. With deepest admiration for my colleagues, I respectfully dissent. * * * “A preliminary injunction is an extraordinary remedy” not to be granted freely.1 The district court should deny relief “unless the party seeking it has clearly carried the burden of persuasion”2 by showing four things: (1) they are likely to succeed on the merits, (2) they are likely to suffer irreparable harm without an injunction, (3) the balance of the equities tips in their favor, and (4) an injunction is in the public interest.3 As for factor one—likelihood of success on the merits—the United States’ claim under § 10 of the Rivers and Harbors Act (RHA) is unlikely to prevail.4 Both the district court and the majority opinion cite inapplicable statutes, treaties, cases, and other evidence to argue that this segment of the Rio Grande is navigable and thus subject to the RHA. But the district court’s erroneous navigability finding—and the majority’s affirmance—cannot be squared with a century-plus of precedent. The United States also falls short on the three other preliminary injunction factors. It is entirely unclear how the preliminary injunction— which orders the barrier to be moved, but not removed (as Mexico demands)—remedies the United States’ diplomatic harms. And as for the balance of the equities and public interest, the United States offers no substantiated record evidence that could justify a preliminary injunction. I The United States has not shown a likelihood of success on the merits of its claim under the first two clauses of RHA§ 10. Under both clauses, the United States must show that this 1,000-foot stretch of the Rio Grande is navigable—but it cannot. More than a century of precedent points to only one conclusion: This 1,000-foot segment is not navigable. The majority opinion and the district court overlook this well-settled law and distort the test for navigability. Despite the majority’s and district court’s many attempts to argue otherwise, the United States has not shown that it will likely prevail in demonstrating that these 1,000 feet were historically used or susceptible of use in interstate or foreign commerce and thus historically navigable.5 The majority does not reach the district court’s alternative holding—that this Rio Grande segment is navigable because reasonable improvements could make it susceptible of future commercial use. But the district court erred there, too: The United States cannot show that it will likely prevail in establishing navigability on this ground because it has failed to present evidence that any improvements would be reasonable.6 A In upholding the district court’s finding of historical navigability, the majority opinion repeats many of the district court’s mistakes—and makes some new ones of its own. 1 Start with the mistakes that the majority opinion and the district court share. Both recite statutes, treaties, a 1975 study by the United States Army Corps of Engineers (the Corps), and cases that they say show historical navigability. But this evidence does not show that this 1,000-foot segment of the Rio Grande was historically used or susceptible of use in commerce. First, the statutes. The district court should not have relied on these statutes to find navigability. In Oklahoma v. Texas, the Supreme Court said in no uncertain terms that a similar act that “provided in substance that there should be no interference with navigation . . . was only precautionary and not intended as an affirmation of navigable capacity in that locality.”7 The statutes cited by the district court and the majority, like the statute in Oklahoma, do not make the necessary factual findings8 that the Rio Grande was used or susceptible of use in commerce; they only prohibit obstructions to navigation.9 In Oklahoma, the Supreme Court explained that this prohibitory language was merely Congress playing it safe by barring obstructions in case that segment of the river turned out to be navigable.10 We should not read these statutes as saying more than they do. Second, the treaties. Like the district court, the majority points to the Treaty of Guadalupe Hidalgo and the Gadsden Treaty, which require the United States and Mexico to maintain “free and common” “navigation of” the Rio Grande below New Mexico. But point is all the majority and district court do. They give no reason—because there is none—why these treaties establish navigability and are not merely precatory. Even a cursory review of the case law shows that these treaties do not establish navigability. In United States v. Rio Grande Dam & Irrigation Company, the Supreme Court said that these treaties obligated the United States “to preserve . . . the navigability of its navigable waters.”11 That is, whether the river is navigable is a preliminary question that is not answered by the treaties themselves. Oklahoma suggests the same. Like the statute in Oklahoma, these treaties do not make specific factual findings of navigability and provide only that there should be no interference with “free and common” navigation.12 On Oklahoma‘s reasoning, then, the treaties are not “affirmation[s] of navigable capacity” but rather “precautionary” statements that navigability should not be obstructed where it exists.13 Because the treaties cannot establish navigability, the majority opinion also improperly relies on the United States Coast Guard’s 1984 navigability determination—which itself relies on these treaties. Third, the Corps’s 1975 study. The majority opinion and the district court cite the Corps’s 1975 study and subsequent determination that the Rio Grande is navigable—but the Corps never found navigability based on historical or then-current use.14 The Corps observed that there was “no [then-current] commercial activity occurring within” that stretch of the river and that there were only “sketchy accounts on [past] use.” The Corps instead found navigability based on the treaties and the Supreme Court’s decision in Rio Grande Dam, which, as I explain, was improper. The majority opinion admits that the Corps’s navigability finding was not based on past use—but it inconspicuously relegates that concession to a footnote and avoids mentioning that the Corps criticized the historical evidence on which it now relies. Like the district court, the majority opinion looks under the hood at the Corps’s historical findings. But the majority’s and district court’s analyses of this historical evidence are flawed. Both overlook a key feature of the test for navigability: Use of the river must have been more than “sporadic,” “ineffective,”15 and “exceptional.”16 Any evidence of past use in the Corps’s study is too sporadic and exceptional to establish historical navigability. The majority opinion and the district court both cite an anecdote that, in 1850, “[a] keelboat and a skiff, manned by sixteen men, ascended the river by channel to a point a thousand miles above the head of steam travel,” which was Roma, Texas. But the district court omitted—and the majority opinion overlooks—that this historical account also described the expedition as “an astonishing penetration for a river with so little water.” And they ignore the Corps’s comment that “[t]here is no showing that substantial items of commerce were shipped from [Roma]” at the time of the expedition. In context, this expedition is at most an “ exceptional” use of the river, which does not suffice to show navigability.17 In addition, the Corps stated that “there apparently has never been any ‘practical navigation’ between Roma . . . and El Paso” and that “at normal stages the river apparently was not navigable above Rio Grande City.” Even “during periods of sufficient flow,” only “fishing boats and other shallow draft craft” could navigate the river. As the Supreme Court explained, navigability does not extend to “every small creek in which a fishing skiff or gunning canoe can be made to float at high water.”18 Rather, to be historically navigable, the river must have been suitable “as [a] highway[] for commerce, over which trade and travel [was] or [could have been] conducted in the customary modes of trade and travel on water.”19 There is no evidence of that here. As the Corps said, any accounts of the river’s use were “sketchy,” and “actual accounts of commercial travel [were] lacking.” And given the “serious ecological objections to any channelization” and “doubtful” economic justifications, there were no reasonable improvements that could have permitted commercial use of that stretch of the river.20 Because the Corps’s study does not show that the river segment was used or susceptible of use in commerce, it does not support the district court’s finding of historical navigability. Fourth, the cases. The majority opinion and the district court cite cases that mention past ferry traffic across the Rio Grande at Eagle Pass. But bank-to-bank traffic alone has never established navigability. Because the test for historical navigability turns on whether the water was used or susceptible of use as a “highway for commerce,” there must also be evidence of commerce or navigation along the river.21 Consistent with this view, the Supreme Court in United States v. Appalachian Electric Power Company did not rely solely on evidence of ten “small public ferries going from one bank to the other” to find navigability.22 Rather, the Court cited ample evidence of boating and commerce along the river, including “[f]ourteen authenticated instances of use . . . by explorers and trappers,” “general historical references to the river as a water route for the early fur traders,” and “evidence of actual use of [one segment] for commerce.”23 Our circuit has taken the same approach. In Puente de Reynosa, S.A. v. City of McAllen, we noted that there was ferrying across the river—but we also observed “that high-pressure steamboats [previously] made frequent trips up [that segment of] the Rio Grande” and that “small boats continue to be navigated on the river” before finding that segment navigable.24 Accordingly, without sufficient evidence of past use or susceptibility to use in commerce along this 1,000-foot Rio Grande segment, there can be no finding of historical navigability. As explained, the “sketchy” historical accounts of activity along the river are nowhere near as extensive as those in Appalachian Electric and do not establish navigability. The majority opinion thus repeats the district court’s mistakes. Both engage only partially with the case law and record, overlooking signs that this stretch of the river was not used or susceptible of use in commerce—and thus not historically navigable. 2 In an attempt to bolster the district court’s erroneous conclusions, the majority opinion points to additional historical anecdotes and the Supreme Court’s opinion in Rio Grande Dam. But in doing so, it makes two new, unforced errors: It would have us ignore Texas geography and give dispositive weight to an out-of-context quote from the Supreme Court. First, the majority opinion quotes historical anecdotes that refer to other parts of the Rio Grande—not this specific 1,000-foot segment near Eagle Pass. Whether the Rio Grande was historically navigable along these other segments has zero bearing here because we analyze navigability of this segment only.25 Consider the majority opinion’s quote from the 1949 book Rio Grande, River of Destiny by Laura Gilpin. Nothing in that quote suggests that this 1,000-foot stretch is within the 200 navigable miles that Gilpin describes. The quote mentions the 200-mile segment and then says that goods were brought to the Rio Grande’s mouth at the Gulf of Mexico, more than two hundred miles from inland Eagle Pass. This suggests that Gilpin’s 200-mile navigable segment started at the Rio Grande’s mouth and thus stopped short of where the barrier is today. The majority’s quote from People and Plots on the Rio Grande suffers from a similar geographic flaw. That quote describes navigation “between [Fort] Ringgold and Brownsville, and to some extent as far up as Laredo.”26 But our 1,000-foot segment is much further inland than Fort Ringgold, Brownsville, and Laredo. These quotes are thus irrelevant. Second, the majority opinion omits the context surrounding the Supreme Court’s comment in Rio Grande Dam that courts may take judicial notice “[t]hat the Rio Grande, speaking generally, is a navigable river.”27 After stripping the qualifying language, the majority says we should give this quote “great, if not dispositive, weight.” The surrounding language in Rio Grande Dam shows that this quote is not persuasive, let alone dispositive. The Supreme Court suggested only that courts could take judicial notice “[t]hat the Rio Grande, speaking generally, is a navigable river” because it is common knowledge.28 The Court was careful to underscore a critical point: “[H]ow far up the stream navigability extends . . . should be determined by evidence” if it is not “a matter of general knowledge, or one that ought to be generally known.”29 The Court added, “[I]t is not so clear that it can fairly be said . . . [that it is] a matter of common knowledge at what particular place between its mouth and its source navigability ceases.”30 In Rio Grande Dam, the Supreme Court tellingly did not take judicial notice that the Rio Grande was navigable within New Mexico—instead, it examined “affidavits and other evidence.”31 And, the real kicker, it concluded that the segment was not navigable.32 Judicial notice of the Rio Grande’s general capacity for navigation clearly does not displace or even supplement our standard fact-driven, segment-by-segment navigability test. As today’s dispute shows, the navigability of this 1,000-foot Rio Grande segment is far from common knowledge. The Supreme Court’s comment about judicial notice thus has no role here. In sum, there is not sufficient evidence to find that the United States is likely to prevail in showing that this specific 1,000-foot stretch of the Rio Grande was historically used or susceptible of use in commerce. The district court clearly erred in finding navigability on this ground. B Because the majority opinion affirms the district court’s finding that this 1,000-foot segment was historically navigable, it does not reach the district court’s alternative holding: that this part of the river is navigable because it could be used for commerce with reasonable improvements.33 I would reject this alternative holding, too. Navigable waters include those that are not presently used as highways for commerce but could be after reasonable improvements.34 Key is the word “reasonable.”35 That is, “[t]here must be a balance between cost and need at a time when the improvement would be useful.”36 The district court said only that future improvements are “possible”—it did not attempt to assess costs and benefits. But how could it? The United States has presented no evidence of costs or commercial benefits.37 The record thus does not support the district court’s holding that the Rio Grande segment is navigable because reasonable improvements could make it suitable for commerce. * * * Because the United States cannot show that this segment is navigable, based on historical evidence or on future reasonable improvements, its RHA claim cannot succeed.38 The United States thus fails to show a likelihood of success on the merits—”the most important of the preliminary injunction factors.”39 II The United States fares no better on the last three preliminary injunctions factors: It has not shown that it “is likely to suffer irreparable harm in the absence of preliminary relief, that the balance of equities tips in [its] favor, and that an injunction is in the public interest.”40 The district court erred in concluding that the United States was likely to suffer irreparable harm without the preliminary injunction. Critically, it is entirely unclear how the injunction alleviates the United States’ diplomatic harms. The record indicates that the alleged harms arise from the “construction and presence” of the barrier and that Mexico has demanded the barrier’s “prompt removal.” The preliminary injunction, however, requires Texas to shift the barrier to the Texas bank, not remove it from the river. If the district court credited the United States’ allegations of harm, then it should have ordered the barrier to be not just moved but removed. Only complete removal would eliminate the “construction and presence” of the barrier and meet Mexico’s demands. The district court’s justification is unsatisfying: It suggested only that ordering removal would not be a “measured” response suitable for the preliminary injunction stage, wholly ignoring the mismatch between the United States’ alleged harms and the remedy it prescribed. That the district court entered a preliminary injunction that is nonresponsive to the United States’ alleged diplomatic harms suggests that those harms are merely speculative.41 The United States also fails to show that the balance of the equities and the public interest favor granting the injunction. The district court erroneously identified three reasons that it said favor the United States: the barrier threatens human life and safety, impairs navigation, and violates the RHA. First, Texas and the district court disagree about the barrier’s safety and usefulness for deterring drug trafficking. Texas argues that the barrier “was designed . . . to save lives and direct migrants to appropriate . . . points of entry while deterring unlawful, dangerous crossings; drug smuggling; human trafficking; and terrorist infiltration” at “one of the most active drug-trafficking and human-trafficking hotspots on the river.” At this stage, however, Texas has not offered concrete evidence that the barrier has saved lives or reduced illegal crossings and drug trafficking. On the other side, the United States—which bears the burden on the preliminary injunction factors42—has offered no credible evidence of harm. Curiously, the district court tried to spin the river’s naturally treacherous conditions as evidence that the barrier is dangerous. The majority opinion also makes that logical leap, but I cannot. To support the district court’s assessment, the majority opinion cites a quote by Governor Abbott that migrants “risk their lives by crossing illegally through the [Rio Grande].” The majority opinion holds this out as evidence that “the floating barrier pose[s] a risk to human life.” But Governor Abbott was talking about the danger posed by the river, not the barrier. The majority opinion cannot misread Governor Abbott’s comment to save the United States on these preliminary injunction factors. The district court also erred in taking judicial notice of news articles that reported that two dead bodies were found near the barrier.43 Although “courts reviewing preliminary injunctions can take judicial notice of subsequent factual developments,”44 those facts must still be proper material for judicial notice under Federal Rule of Evidence 201.45 Under that rule, a district court “may judicially notice a fact that is not subject to reasonable dispute because it: (1) is generally known within the trial court’s territorial jurisdiction; or (2) can be accurately and readily determined from sources whose accuracy cannot reasonably be questioned.”46 The news articles satisfy neither prong: It was not established that the facts were “generally known within the” district court’s jurisdiction, and Texas reasonably questioned the articles’ accuracy.47 The district court should not have found on the evidence before it that the barrier is a “threat to human life.” Second, the district court should not have found that the barrier is an “impairment to free and safe navigation” and a “contraindication to the balance of priorities Congress struck in the RHA.” Because this Rio Grande segment is not navigable and the United States is likely to fail on its RHA claim, neither of these factors carries weight. In sum, the district court’s analysis of these equitable factors was unpersuasive, unsubstantiated, and incorrect—and the majority opinion repeats the error. The United States has not carried its burden. III A preliminary injunction is an exceptional remedy that requires the moving party to clearly satisfy all four requirements. In this case, the United States cannot satisfy one, much less all four. The law and the record are clear: The United States cannot succeed on its RHA claim because it has not shown that this 1,000-foot segment of the Rio Grande is navigable. In reaching the opposite conclusion, the majority opinion and the district court resort to evidence that is foreclosed to us by a century-plus of case law. Nor can the United States satisfy the three other preliminary injunction factors. As the United States has not “clearly carried the burden of persuasion”48 on even one requirement to obtain the “extraordinary remedy”49 of a preliminary injunction, I respectfully dissent.
