¶ 1 Carole Bishop and Mark Klosky ("Klosky") 1 and Shannon and Keith Love ("the Loves") own adjacent parcels of land in Denver's Washington Park neighborhood. Klosky wants to remove a large tree sitting primarily on Klosky's property and partly on the Loves' property. The Loves want to keep the tree. Our seemingly straightforward task is to decide what analytical framework should govern this garden-variety dispute.

¶ 2 In undertaking this task, we do not write on a clean slate. Rhodig v. Keck holds that when a tree encroaches onto a neighbor's land, the tree remains the sole property of the owner of the land where the tree first grew, unless the tree was jointly planted, jointly cared for, or treated as a partition between the two properties. 161 Colo. 337 , 421 P.2d 729 , 731 (1966). Any such joint activity implies a shared property interest. Id. Applying Rhodig , the lower courts determined that the Loves cannot prevent Klosky from removing the tree because the Loves have failed to prove any such shared property interest in the tree.

¶ 3 If we're unwilling to construe Rhodig in their favor, the Loves ask us to overrule it and simply hold that a tree becomes joint property whenever it crosses a property line. We decline the Loves' invitation in both respects. As we explain more fully below, we do not construe our precedent in their favor. And we conclude that there is no sound legal basis for abandoning Rhodig . After surveying the early common law on which Rhodig is premised, we instead clarify that Rhodig governs "encroachment trees," which are those that begin life entirely on one person's property only to migrate partially to another's. Under Rhodig , a landowner may remove such a tree without first securing the approval of his neighbor, unless the landowners jointly planted, jointly cared for, or treated the trees as a partition between the properties. Here, the Loves did not prove such joint activity implying shared ownership of the encroaching tree. So, Klosky may remove the tree.

¶ 4 Thus, we affirm the judgment of the court of appeals and remand for further proceedings consistent with this opinion.

I. Facts and Procedural History

¶ 5 In central Denver's Washington Park neighborhood a seventy-foot-tall catalpa tree towers over two adjacent properties. At the base of its trunk, the tree sits roughly three- *1269 quarters on Klosky's property and one-quarter on the Loves' property. Estimated to be between seventy and ninety years old, the tree inhabited the lots well before the parties moved in, and it is unknown who planted it (or whether it might have simply sprung up as a "volunteer"). What we do know is that the tree sheds leaves, seed pods, and branches on both properties. 2

¶ 6 Unhappy with the debris, Klosky wants to cut the tree down. The Loves tried to convince Klosky otherwise, but to no avail. Therefore, the Loves filed a lawsuit seeking a temporary restraining order ("TRO"), declaratory judgment, and a preliminary injunction to prevent Klosky from felling the tree. The parties agreed to a TRO pending the preliminary injunction hearing, and also agreed to treat that hearing as a trial on the merits.

¶ 7 The trial court ruled in favor of Klosky, articulating what it perceived as the test for joint ownership of a boundary tree in Rhodig : "In addition to proving it straddles the line, ... [the Loves have] to prove that the tree was jointly planted by the property owners or their predecessors, that it was jointly maintained, or that it was jointly treated as the boundary."

¶ 8 The trial court made the following factual findings:

• 74% of the tree trunk at its base is on Klosky's property and 26% of the trunk at its base is on the Loves' property;

• the tree started life on Klosky's land and grew partly onto the Loves' land;

• the parties did not plant the tree;

• the parties did not jointly maintain the tree in a way that might reflect "an implicit understanding that the tree is jointly owned"; and

• the parties did not intend the tree to serve as a boundary.

¶ 9 Based on these findings, the trial court concluded that the Loves had not met their burden of proving that they had a property interest in the tree. So, it dismissed the Loves' claims and dissolved the TRO, but then stayed those orders pending appeal.

¶ 10 On appeal, the Loves contended that (1) the trial court erred in concluding Klosky and the Loves did not jointly care for the tree and (2) Rhodig should be overturned. A division of the court of appeals affirmed the trial court in a published opinion, Love v. Klosky , 2016 COA 131 , ¶ 3, ---P.3d ----. Because the Loves failed to designate the relevant trial testimony as part of the record on appeal, the division determined it had insufficient information to review the trial court's decision regarding joint care. Id. at ¶ 12. The division held it was bound by Rhodig , which it interpreted to mean that "boundary trees are held as common property only if the landowners jointly planted, jointly cared for, or treated the trees as a partition between the properties." Id. at ¶ 20. The division also declined to address the Loves' argument, raised for the first time in the Loves' reply brief, that Rhodig 's test was intended to apply only to trees planted by trespass. See id. at ¶ 11. Two members of the division called for this court to overturn Rhodig and require instead that any time a tree straddles two lands, the adjacent property owners jointly own the tree as tenants-in-common. Id. at ¶ 29.

¶ 11 The Loves sought our review, and we granted certiorari to decide whether to overrule Rhodig . 3

II. Analysis

¶ 12 We begin by identifying the standard of review, and then we summarize the doctrine of stare decisis, which governs when we should deviate from prior case law. We then consider whether to abandon Rhodig . To answer that question, we first evaluate what Rhodig means. We clarify that Rhodig outlines the test for determining ownership of an encroachment tree. We conclude that *1270 Rhodig has no bearing on the common law test for so-called "true boundary line" cases, that is, those cases that involve a tree straddling a property line without evidence of migration. Finally, we explain why the Rhodig rule, properly understood, remains sound for encroachment cases like the one before us here. Lacking a sound legal basis for abandoning our long-standing precedent in Rhodig , we conclude that we should not overrule it.

A. Standard of Review

¶ 13 We review questions of law de novo. People v. Porter , 2015 CO 34 , ¶ 8, 348 P.3d 922 , 924.

B. Stare Decisis

¶ 14 Stare decisis is a judge-made doctrine that requires courts to follow preexisting rules of law. Id. at ¶ 23, 348 P.3d at 927 . "Stare decisis is the preferred course because it promotes the evenhanded, predictable, and consistent development of legal principles, fosters reliance on judicial decisions, and contributes to the actual and perceived integrity of the judicial process." Payne v. Tennessee , 501 U.S. 808 , 827, 111 S.Ct. 2597 , 115 L.Ed.2d 720 (1991) ; see also People v. LaRosa , 2013 CO 2 , ¶ 28, 293 P.3d 567 , 574. Because of these virtues, courts are reluctant to undo settled law. Creacy v. Indus. Comm'n , 148 Colo. 429 , 366 P.2d 384 , 386 (1961).

¶ 15 Still, courts may depart from a prior ruling, or overrule it, where sound reasons exist to do so. Nicholls v. People , 2017 CO 71 , ¶ 29, 396 P.3d 675 , 681. We will depart from our existing law only if we are clearly convinced that (1) the rule was originally erroneous or is no longer sound because of changing conditions and (2) more good than harm will come from departing from precedent. McShane v. Stirling Ranch Prop. Owners Ass'n , 2017 CO 38 , ¶ 26, 393 P.3d 978 , 984.

¶ 16 With these principles in mind, we turn to Rhodig .

C. The Rhodig Rule Embraces the Masters Approach for Encroachment Cases

¶ 17 Before deciding whether Rhodig should be overruled, we must first understand what Rhodig means. In order to understand what it means, it is helpful to evaluate the two prevailing approaches for determining the property rights of trees that have encroached on another's property. We can then see where Rhodig sits in historical context and what some of its more cryptic passages seek to convey.

¶ 18 While the division below saw Rhodig as part of the minority of cases rejecting what the division perceived as the common law majority rule for boundary cases generally (namely, that any tree straddling a boundary is joint property, Love , ¶¶ 22-23 ), we view the relevant split of authority somewhat differently. The common law rule regarding true boundary-line-tree cases, where the tree sits squarely on a property boundary with no evidence of migration, is not implicated here. See Dubois v. Beaver , 25 N.Y. 123 , 126 (1862) (describing boundary-line trees as "trees deriving their nourishment from roots extending on both sides of the line, and with bodies so directly over the line, and necessarily on both sides of that line, that it could not be determined upon which side of the line the tree was originally planted"). In those cases, the tree is the joint property of the two adjacent property owners. See 1 Am. Jur. 2d Adjoining Landowners § 17 (2018) ("[A] tree standing on the division line between adjoining landowners is generally considered the common property of both landowners...."). This is not such a case. Instead, this case involves a tree from one property that grew and encroached on another. History reveals two approaches to determining ownership in these encroachment cases. We turn to those two approaches now.

1. The Masters and Waterman Approaches to Encroachment Cases

¶ 19 In 1620, Masters v. Pollie announced that when the roots of a tree planted in the land of one owner spread to a neighbor's *1271 land, the tree belonged to the owner of the land in which the tree was planted. Roscoe Pound, Juristic Science and Law , 31 Harv. L. Rev. 1047 , 1049 (1918) (citing Masters , (1620) 81 Eng. Rep. 712 (K.B.) 712; 2 Roll. Rep. 141, 141).

¶ 20 In contrast, Waterman v. Soper held that if a tree on A's land extended its root onto the land of B, then A and B became tenants-in-common. (1697-98) 91 Eng. Rep. 1393 (K.B.) 1393; 1 Ld. Raym. 737, 738 ("A tree belongs to the person or persons in whose land the root grows.").

¶ 21 In Holder v. Coates , the British court resolved this split in favor of Masters . (1827) 173 Eng. Rep. 1099 (K.B.) 1100; 1 Moody & M. 112, 113. There, the trunk of the tree was on the defendant's land, "but some of the lateral or spur roots grew into the land of both parties." Id. at 1099 ; 1 Moody & M. at 112. The Holder court recited the differing approaches in Masters and Waterman , ultimately holding that if the jury could ascertain on which property the tree was planted, the jury should find in favor of that property owner. Id. at 1100 ; 1 Moody & M. at 113-14.

¶ 22 Early American cases recognized the Masters / Waterman split. For example, in Dubois v. Beaver , citing Masters and Waterman , the court outlined the encroachment-tree split. 25 N.Y. at 126-28 . However, the Dubois court did not select a preferred approach, because it held that the plaintiff would have prevailed had he owned the tree alone or jointly with the defendant. Id. at 128 .

¶ 23 In modern American times, debate continues about how to treat tree ownership. Even among jurisdictions that follow Masters and recognize that some trees with parts crossing a boundary are encroaching trees that are not jointly owned, another variation of the debate has sprouted: How do you tell the difference between boundary-line trees and encroaching trees? Some jurisdictions treat any tree with part of the trunk on both sides as a boundary tree. See, e.g. , Blalock v. Atwood , 154 Ky. 394 , 157 S.W. 694 , 694-96 (Ky. Ct. App. 1913) (applying the rule that a tree is common property even though the tree was three-quarters on one property and one-quarter on the other property); Ridge v. Blaha , 166 Ill.App.3d 662 , 117 Ill.Dec. 629 , 520 N.E.2d 980 , 981-83 (1988) (holding property owners to be tenants-in-common even where the tree was largely on one property). Others would treat trees that started on one side and extended parts-even parts of their trunks-to the other, as encroaching trees. See, e.g. , Holmberg v. Bergin , 285 Minn. 250 , 172 N.W.2d 739 , 743 (1969) (requiring more than the presence of the trunk across the boundary line to establish joint ownership); Garcia v. Sanchez , 108 N.M. 388 , 772 P.2d 1311 , 1314 (Ct. App. 1989) (citing Holmberg in approval and affirming the trial court's determination that the encroaching trees were not commonly owned by the two property owners).

2. Rhodig v. Keck

¶ 24 We joined the encroachment-tree debate in Rhodig v. Keck , where the Rhodigs sued their neighbor, Roy Keck, for damages because Keck had removed four trees near their shared property line. 421 P.2d at 730 . When the Rhodigs purchased their property, there were two such trees. Id. The Rhodigs planted two more, and later replaced one of the original two trees after it died. Id. Keck removed all four trees after a survey of the lots showed one tree was entirely on Keck's land and three were primarily on Keck's land and partly on the Rhodigs' land (the opinion does not make clear whether the one tree entirely on Keck's land was the one remaining original tree or was one of the three trees the Rhodigs planted). Id. The court determined that "the trees in question, when planted, must necessarily have been wholly upon Keck's property." Id. We then somewhat tentatively outlined a test to see if the Rhodigs had an interest in the trees: "Apparently a test in determining whether trees are boundary line subjects entitled to protection is whether they were planted jointly, or jointly cared for, or were treated as a partition between adjoining properties." Id. at 731. We concluded that the Rhodigs did not prove such joint ownership, and thus, Keck had the right to "remove the encroachment." Id.

¶ 25 The Loves first argue that Rhodig embraced a rule that would classify any tree on the property line as joint property of the *1272 two property owners, but with an exception for trees planted by trespass on a neighboring property. 4 We are unpersuaded by their argument for two reasons. First, nothing in Rhodig limits its test to trees planted by trespass. Rhodig prefaces its test by referring to "trees," with no mention of whether the trees were planted by trespass. Id. ("Apparently a test in determining whether trees are boundary line subjects entitled to protection is whether they were planted jointly, or jointly cared for, or were treated as a partition between adjoining properties." (emphasis added)). Second, the court applied the same test to all trees in question, even though the record revealed that only three of the four trees were necessarily planted by trespass. Id. The court did not apply the test to the one tree completely on Keck's land. Id. ("Obviously here one of the trees being wholly on Keck's land, it is not involved in this dispute under these facts."). Thus, we conclude Rhodig did not merely create an exception for trees planted by trespass.

¶ 26 The Loves argue in the alternative that we should overrule Rhodig and adopt an approach that would automatically make the Loves tenants-in-common with Klosky because the tree crossed the property line.

¶ 27 Klosky, on the other hand, argues that Rhodig holds that when a tree crosses over a boundary it remains the property of the owner of the land on which the tree originally grew, unless one of the joint-action situations enumerated in Rhodig applies. We agree with the outcome Klosky suggests, although we take a different path in reaching it.

¶ 28 We surmise that our ambiguous precedent caused the lower courts to conflate the common law rule for true boundary-line cases and the test for encroachment trees. The lower courts both described Rhodig as rejecting the common law rule for boundary-line cases and as creating a test that applies to all trees on a property line, not just encroachment cases. The district court articulated the test as applying to a tree that "straddles the line." Likewise, the division described the rule as "a rule that governed all boundary trees." Love , ¶ 20.

¶ 29 Instead, we infer that Rhodig embraced the Masters approach for encroachment trees; and in so doing, the Rhodig court defined encroachment trees as trees that start life on one property and grow onto a neighboring property. See 421 P.2d at 730-31 . The court specifically noted that Rhodig involved an "encroachment," id. at 731, which it differentiated from a "true boundary line case": "Based on the factual situation presented, plaintiffs' contention that they and Keck were tenants in common of the trees is without merit, for this is not a true boundary line case. The line of authorities relied on by the Rhodigs, therefore, is not in point." Id. at 730 (citations omitted). Furthermore, the cases Rhodig lists as inapplicable are all cases where the trees were described generally as being on the border of the property, and there was no suggestion the tree was planted on one property and encroached onto another. See, e.g. , Scarborough v. Woodill , 7 Cal.App. 39 , 93 P. 383 , 383 (1907) ("Plaintiff and defendant own adjoining orange orchards in the county of Riverside, which are separated by a row of cypress trees growing on the boundary line between them." (emphasis added)); Musch v. Burkhart , 83 Iowa 301 , 48 N.W. 1025 , 1025 (1891) (describing the trees in question as ones connecting a wire fence and standing "on the common boundary line"); Cathcart v. Malone , 33 Tenn.App. 93 , 229 S.W.2d 157 , 158 (1950) ("The trees in question as shown by the proof and especially the exhibits filed were on the line, partly on property of each of the parties ." (emphasis added)). Thus, the majority's holding in Rhodig has no bearing on cases involving "true boundary line" trees or where it is impossible to discern on which property the tree began. The Rhodig dissent, in contrast, cites to Waterman , suggesting the wiser choice would have been to make the Rhodigs and Keck co-owners of the trees. 421 P.2d at 731 (Frantz, J., dissenting).

*1273 ¶ 30 Rather than change the rule for all boundary trees, Rhodig merely chose the Masters approach for distinguishing encroachment trees from boundary-line trees. While adding caveats for when an encroachment tree could become joint property-namely, when the tree was jointly planted, jointly cared for, or treated as a partition between the properties- Rhodig embraces Masters . A tree does not automatically become a boundary-line tree, and thus joint property, merely by touching a property line.

3. Rhodig Remains Sound

¶ 31 Having clarified what Rhodig means, we now return to the doctrine of stare decisis and consider whether we are clearly convinced that (1) Rhodig was originally erroneous or is no longer sound because of changed conditions and (2) more good than harm will come from departing from precedent. See McShane , ¶ 26, 393 P.3d at 984 . We find neither prong of this standard is met here.

¶ 32 First, Rhodig was correctly decided. We know of no other context in which a transfer of real property occurs with no action on behalf of either party and with no intent to transfer the property interest. Cf. § 38-35-101, C.R.S. (2017) ("Any deed or other instrument relating to or affecting title to real property acknowledged substantially in accordance with the following form before a proper official shall be prima facie evidence of the proper execution thereof...."). We are loathe to create a rule allowing for the automatic transfer of a property interest because an encroaching tree touches a property line.

¶ 33 While this is a case where the encroached-upon neighbor wants to keep the tree, typically, in similar disputes, the encroached-upon neighbor would be attempting to remove the encroaching tree. But under the Waterman approach (or the common law boundary-tree approach advocated by the Loves), such an encroached-upon neighbor would have no recourse to remove the tree over the objection of her neighbor because both neighbors would be required to consent to the tree's removal.

¶ 34 We can imagine many situations where the encroached-upon neighbor becomes an unwilling joint owner of the tree. Consider a tree that encroaches from A's land onto neighbor B's land and the roots begin pushing into B's sewage lines. Under the Rhodig / Masters rule, B could possibly bring a nuisance action against A for unreasonably interfering (via A's tree) with B's property. See Pub. Serv. Co. of Colo. v. Van Wyk , 27 P.3d 377 , 391 (Colo. 2001) ("To maintain a successful nuisance claim, a plaintiff must establish that the defendant has unreasonably interfered with the use and enjoyment of her property."). Yet, under the Waterman approach, if A wants to keep the tree, B could not bring a nuisance action because B would be the tree's joint owner (despite taking no action, and having no intent, to become a joint owner). No theory of nuisance liability would provide relief because B's property cannot invade itself. See Hoery v. United States , 64 P.3d 214 , 218 (Colo. 2003) ("Liability for nuisance may rest upon any one of three types of conduct: an intentional invasion of a person's interest; a negligent invasion of a person's interest; or, conduct so dangerous to life or property and so abnormal or out-of-place in its surroundings as to fall within the principles of strict liability.").

¶ 35 Or perhaps the tree barely inches over to B's land. B doesn't notice the minor intrusion onto her land. Under the Waterman approach, now B not only has the benefit of jointly owning the tree, but also the burden of legal responsibility for the tree. Imagine a branch of the tree falls and injures someone on the sidewalk. Should we now make B jointly liable for a tree that she-and anyone else who had not conducted an exacting land survey recently-thinks is her neighbor's property? Cf. § 13-21-115, C.R.S. (2017) (describing the circumstances under which a civil action may be brought against landowners by injured persons).

¶ 36 Thus, while this is a case where the encroached-upon neighbor wants to keep the tree, adopting the Waterman approach here would also have the consequence of imposing that same rule upon all other encroached- *1274 upon neighbors, many of whom may not want to be joint owners of an encroaching tree.

¶ 37 Because the Rhodig approach does not automatically transfer a property interest in encroachment-tree cases, we conclude that it remains sound. And, we see no conditions that have changed to make the above reasoning any less compelling today than when we decided Rhodig .

¶ 38 Finally, we do not see how overruling Rhodig to adopt the Waterman approach (or the common law boundary approach advocated by the Loves), would do more good than harm. As described above, there are many potential problems with such a change. And adopting the Waterman approach would simply switch winners in this case. As Judge Dailey noted in his special concurrence, "Changing winners is not a sufficient reason for overruling prior precedent." Love , ¶ 37 (Dailey, J., specially concurring).

¶ 39 Therefore, we decline the invitation to overturn Rhodig .

D. The Loves Have Not Demonstrated Joint Ownership Under Rhodig

¶ 40 Now that we've clarified and upheld Rhodig , we apply it to resolve the dispute at hand.

¶ 41 First, the tree here is an encroachment tree. Because the trees in Rhodig , "when planted, must necessarily have been wholly upon Keck's property," they were not "true boundary" trees. 421 P.2d at 730 . The same is true here-the trial court found that the tree first grew on Klosky's land and later encroached onto the Loves' land.

¶ 42 Second, the Loves have not sufficiently shown other circumstances that could create joint ownership of the encroaching tree. Just as the Rhodigs had no property interest in the trees that had encroached onto their land because there was not sufficient evidence the parties jointly planted the trees, jointly cared for the trees, or intended for the trees to serve as a boundary, here, the Loves have no property interest in the tree that has encroached onto their land because they have not shown such joint activity implying shared ownership.

III. Conclusion

¶ 43 Because the tree here is an encroachment tree, Rhodig controls. And, under Rhodig , a landowner may remove a tree on his property that grew onto his neighbor's land without first securing the approval of his neighbor, unless the landowners jointly planted, jointly cared for, or treated the tree as a partition between the properties. Here, the Loves did not prove such joint activity implying shared ownership of the encroaching tree. So, Klosky may remove the tree.

¶ 44 Thus, we affirm the judgment of the court of appeals, albeit on different grounds, and remand for further proceedings consistent with this opinion.

JUSTICE GABRIEL does not participate.