OPINION
This case is a continuation of the general adjudication of the Rio Hondo River system. It specifically involves the water rights of the Mescalero Apache Indian Reservation. See State ex rel. Reynolds v. Lewis, 88 N.M. 636, 637, 545 P.2d 1014, 1015 (1976) [hereinafter Lewis I], in which our Supreme Court held that the McCarran Amendment, 43 U.S.C. § 666(a) (1970), granted state courts jurisdiction over the United States as owner, in a trust capacity, of the Indians’ water rights when state courts are conducting general stream adjudications. The issues not addressed in Lewis I, i.e., the extent of the Mescalero Tribe’s water rights and the measure by which they should be determined, were adjudicated by the court below. Specifically, that court ruled that the United States on behalf of the Tribe was entitled to a diver*197sion of 2322.4 acre-feet per year with a priority date of no earlier than 1873, the date of the first executive order delineating the boundaries of the Mescalero Apache reservation.
In contrast, the United States and the Tribe contended that they were entitled to a diversion of 17,750.4 acre-feet per year with a priority date of time immemorial based on an aboriginal water right or, in the alternative, at least a priority date of 1852, based on a federally reserved water right, pursuant to the treaty between the Apache and the United States, in which the latter promised to establish a reservation for the former. The United States and the Tribe appeal.
For convenience, when we refer to Appellants or to the Tribe in its role as appellant, we intend to refer to both the United States and the Tribe. The Appellees are the State of New Mexico on the relation of the state engineer and various cities, villages, counties, acequia associations, and individual downstream land owners who are members of or represented by the Water Defense Association. Again for convenience, we will refer to them as Appellees or the Water Defense Association. In addition to the Tribe’s appeal, the State has cross-appealed. We set out in some detail the issues on appeal and cross-appeal.
The Tribe’s first issue is that it is entitled to a priority date of time immemorial for its water right based on its aboriginal title to the reservation and all things within it. The Tribe’s second issue is that it is entitled to a priority date of 1852 based on the date of the treaty. The Tribe’s third issue is that it is entitled to 15,428 acre-feet of water beyond what the trial court awarded it, because Indian water rights based on a federal reservation are to be measured by the standard of “practicably irrigable acreage” (PIA) and because Appellants proved they could practicably irrigate enough acreage to result in an additional diversion right of 15,428 acre-feet.
The State’s cross-appeal raises two issues. First, the State challenges the trial court’s decision to use a PIA analysis rather than an analysis that would afford the Tribe their minimal needs or a moderate living. However, the State indicates that its challenge to the use of the PIA standard need not be addressed if we affirm the trial court’s ruling rejecting the Tribe’s request for additional water rights under the PIA analysis. Second, the State contends that the trial court erred in failing to impose a consumptive use cap on the 2322.4 acre-feet of water awarded, which would result in the lowering of the award to 1224.7 acre-feet.
We first announce our ruling and, in doing so, summarily dispose of the Tribe’s first issue and both issues on the State’s cross-appeal. We next dispose of a miscellaneous matter involving a motion filed during the pendency of the appeal. We finally address the Tribe’s second and third issues at length. We reverse the trial court’s setting of the 1873 priority date and hold that the priority date should have been 1852, the date of the treaty. We affirm the trial court’s PIA analysis.
At oral argument, the Tribe conceded that it was seeking no practical relief from its issue arguing for an aboriginal water right with a time-immemorial priority. Specifically, the Tribe conceded that it was not seeking any different or greater quantification by seeking an aboriginal right, and that the difference between a priority date of 1852 and an earlier priority date would afford no practical relief because the area was not settled by non-Indians until after 1852. Thus, an 1852 priority date would establish the Tribe as the most senior water right holder.
Because we will not issue advisory opinions, Behles v. New Mexico Pub. Serv. Comm’n (In re Timberon Water Co.), 114 N.M. 154, 162, 836 P.2d 73, 81 (1992), we do not address the Tribe’s time-immemorial-priority issue or the State’s quantification issue on cross-appeal. The Tribe’s concession and our decision on its 1852-priority issue has made moot its time-immemorial-priority issue. Our decision on the Tribe’s PIA issue has made moot the State’s quantification issue. As for the State’s consumptive-use-cap issue on cross-appeal, the *198State concedes that it did not raise this issue below until after the judgment was entered. Its request for findings and conclusions on the consumptive use cap at that time came too late to preserve this issue for appeal. See Hidalgo v. Córtese (In re Guardianship of Caffo), 69 N.M. 320, 323, 366 P.2d 848, 850 (1961); American Bank of Commerce v. United States Fidelity & Guar. Co., 85 N.M. 478, 513 P.2d 1260 (1973).
We deny the Tribe’s motion to strike the statement of the real parties in interest and enjoin participation by non-parties. It is not clear to us what practical relief would be afforded by granting the motion, and the motion appears to us to be technical niggling. The briefs had already been filed at the time the motion was filed, the alignment of parties and the representation had already been fixed, and oral argument was had in this case with only the attorneys who had filed the briefs (or their substitutes) participating. We do not see what purpose would be served by granting the Tribe’s motion.
PRIORITY DATE BASED ON TREATY OR EXECUTIVE ORDERS
The trial court ruled that the 1852 treaty was a “peace and amity” treaty that expressly did not designate a reservation of land. It found that numerous acts by federal government officials between 1852 and 1873 established that the 1852 treaty did not create the reservation. It found that the reservation was created by five executive orders, dated 1873, 1874, 1875, 1882, and 1883. The trial court found that the first three executive orders drew boundaries with the specific intent of excluding arable lands along the river valleys that had been settled by non-Indian farmers pri- or to 1873. The trial court found that the fourth order deleted some of the reservation land and returned it to the public domain either in response to non-Indian mining interests or because it was occupied by non-Indian settlers. Based on these findings, the trial court concluded that the 1852 treaty did not create the reservation and, accordingly, the treaty could not be used as the basis for a federally reserved water right. It also concluded that the series of executive orders did create the reservation and, therefore, formed the basis of the Tribe’s water right such that the water right attributable to each sector of land would be prioritized from the date of the executive order that first withdrew that particular land for the reservation.
In Lewis I, our Supreme Court agreed with the Arizona v. California, 373 U.S. 546, 600, 83 S.Ct. 1468, 1497, 10 L.Ed.2d 542 (1963) [hereinafter Arizona v. California I], description of federally reserved water rights originally outlined in Winters v. United States, 207 U.S. 564, 28 S.Ct. 207, 52 L.Ed. 340 (1908):
The Court in Winters concluded that the Government, when it created that Indian Reservation, intended to deal fairly with the Indians by reserving for them the waters without which their lands would have been useless____ We follow it now and agree that the United States did reserve the water rights for the Indians effective as of the time that Indian Reservations were created.
Lewis I, 88 N.M. at 639, 545 P.2d at 1017 (emphasis omitted and added). The question we must answer in this case is whether the trial court was correct in determining that the reservation was not created for the purpose of establishing a priority date for water rights at the time of the treaty. While this matter is not entirely free from doubt, we believe that the applicable rules of law support the conclusion that the trial court was incorrect in determining that the Tribe’s water rights did not date from 1852.
In addition to reading the pertinent documents to confirm an aboriginal water right under United States v. Winans, 198 U.S. 371, 25 S.Ct. 662, 49 L.Ed. 1089 (1905), which reading we do not discuss for the reason of mootness described above, there are three ways of reading the pertinent documents to establish a priority date: (1) one can read the treaty as establishing the reservation, in which case the priority date is 1852; (2) one can read the treaty together with the executive orders to establish the reservation, in which case the priority *199date could be 1852; or (3) one can read the treaty, as the trial court did, as a peace and amity treaty, so that the executive orders alone created the reservation, in which case the priority date would be the date of each executive order. In determining which reading to adopt, we are guided by a host of cases, none of which squarely answers our questions. In fact, so little do the cases answer our questions that, frequently, both sides will rely on different portions of the same opinion as supporting their contentions.
In the area of Indian law, it is not unusual for cases to contain dicta, unnecessary to the decision at hand, about historical data or the effect of certain events. In later cases, the courts find it necessary to retreat from that language. Thus, for example, in Northwestern Bands of Shoshone Indians v. United States, 324 U.S. 335, 350 n. 8, 65 S.Ct. 690, 698 n. 8, 89 L.Ed. 985 (1945), the Court recognized that prior casual references to Indian title created or confirmed by a certain treaty were not entitled to much weight because the question was not at issue in those cases.
The parties in this case rely on a series of relatively recent Tenth Circuit Court of Appeals and United States Supreme Court cases involving the 1852 treaty as supporting their positions as a matter of law. However, we do not find these cases to be very persuasive. These cases involved liquor laws, United States v. New Mexico, 590 F.2d 323 (10th Cir.1978), cert. denied, 444 U.S. 832, 100 S.Ct. 63, 62 L.Ed.2d 42 (1979), overruled on other grounds by Citizen Band Potawatomi Indian Tribe v. Oklahoma Tax Commission, 975 F.2d 1459 (10th Cir.1992), state tax on the gross receipts of the contractors who built a resort complex on a reservation, Mescalero Apache Tribe v. O’Cheskey, 625 F.2d 967 (10th Cir.1980) (en banc), cert. denied, 450 U.S. 959, 101 S.Ct. 1417, 67 L.Ed.2d 383 (1981), and hunting laws, Mescalero Apache Tribe v. New Mexico, 630 F.2d 724 (10th Cir.1980), aff'd, 462 U.S. 324, 103 S.Ct. 2378, 76 L.Ed.2d 611 (1983).
In the liquor law case, by way of introduction, the court said that the Tribe was formally placed under government control in the 1852 treaty and that, by the treaty, the Tribe subjected itself to United States laws, and its lands would be held in trust by the United States. United States v. New Mexico, 590 F.2d at 325. In the tax case, the court said that the treaty did not help either party at all because it was only a treaty of peace and it established no rights whatsoever for anyone. O’Cheskey, 625 F.2d at 971. The court also said, however, that that case did not involve Indian lands. Id., at 969. In the hunting case, the two judges who formed the majority relied on the treaty to establish Tribal control over hunting and fishing on the reservation. Mescalero Apache Tribe v. New Mexico, 630 F.2d at 730-31. These two judges incidentally were the dissenters in the en banc tax case, and they relied on the treaty there as an integral part of their reasoning. In a footnote, however, the hunting case recognizes that the State did not challenge the validity of the treaty. Id., at 731 n. 13. Similarly, the Supreme Court, in affirming the hunting case, noted that the State conceded that the Tribe retained sovereignty under the 1852 treaty and that sovereignty included the right to regulate the resources and lands of its reservation, including wildlife. New Mexico v. Mescalero Apache Tribe, 462 U.S. at 337, 103 S.Ct. at 2388. We note that the introduction to that same case states that the reservation was created by the executive orders. Id. at 325-26, 103 S.Ct. at 2381-82.
It is apparent that none of these cases may be used to determine definitively when the Mescalero reservation was created. Reciting what the cases have involved and what they have said demonstrates the problems inherent in this case. These prior cases involving the Mescalero Apache Tribe, while perhaps giving guidance in this case, do not establish as a matter of law when the reservation was created or from what date the Tribe will have priority.
To determine when the reservation was created for purposes of establishing the priority date for the Tribe’s water rights requires an analysis of the principles *200governing construction of documents involving Indians and application of those principles to this case. Those principles include canons of construction that generally favor the Indians. Thus, the cases indicate that treaties are construed more liberally than private agreements; they are to be construed, as far as possible, in the sense the Indians understood them, and in a spirit that generously recognizes the full obligation of the United States to protect its dependents. Choctaw Nation of Indians v. United States, 318 U.S. 423, 432, 63 S.Ct. 672, 678, 87 L.Ed. 877 (1943); see also Washington v. Washington State Commercial Passenger Fishing Vessel Ass’n, 443 U.S. 658, 676, 99 S.Ct. 3055, 3069-70, 61 L.Ed.2d 823 (1979) (treaties are construed as Indians would have understood them); Winans, 198 U.S. at 380-81, 25 S.Ct. at 664 (same); Jones v. Meehan, 175 U.S. 1, 11, 20 S.Ct. 1, 5, 44 L.Ed. 49 (1899) (same); Alaska Pac. Fisheries v. United States, 248 U.S. 78, 89, 39 S.Ct. 40, 42, 63 L.Ed. 138 (1918) (statutes passed for the benefit of Indians are construed liberally in favor of Indians); United States v. Walker River Irrigation Dist., 104 F.2d 334, 337 (9th Cir.1939) (liberal construction of treaties and executive orders in favor of Indians because of their dependent status).
On the other hand, treaties and like documents are not to be construed in such a manner as to rewrite them, and they are not to be expanded beyond their clear terms, even to remedy injustice. Choctaw Nation of Indians v. United States, 318 U.S. at 432; 63 S.Ct. at 678; see also Northwestern Bands of Shoshone Indians v. United States, 324 U.S. at 353, 65 S.Ct. at 699 (treaties are to be construed according to their tenor; generosity toward Indians is for Congress, not the courts); Confederated Bands of Ute Indians v. United States, 330 U.S. 169, 179, 67 S.Ct. 650, 655, 91 L.Ed. 823 (1947) (courts cannot change meaning of treaty under the guise of liberal interpretation; how Indians understood the treaty, if not consistent with language, does not control); United States v. Minnesota, 466 F.Supp. 1382, 1385 (D.Minn.1979) (courts cannot remake history or expand treaties beyond their clear terms, even to remedy perceived injustice suffered by Indians), aff'd sub nom. Red Lake Band of Chippewa Indians v. Minnesota, 614 F.2d 1161 (8th Cir.), cert. denied, 449 U.S. 905, 101 S.Ct. 279, 66 L.Ed.2d 136 (1980).
In interpreting documents involving Indians and the government, the courts are guided by the above principles of construction in considering the documents’ wording. However, the courts also consider the prior history between the parties leading up to the document, the surrounding circumstances, and the construction the parties subsequently adopted as evidenced by their actions. Id.; see also Choctaw Nation of Indians v. United States, 318 U.S. at 431-32, 63 S.Ct. at 678 (to ascertain document’s meaning, one looks to the history, the negotiations, and the practical construction adopted by the parties).
As we view the treaty and the executive orders at issue in this case, the first thing that is apparent to us is that their meaning is not clear. Thus, we do not believe that cases such as Oregon Department of Fish & Wildlife v. Klamath Indian Tribe, 473 U.S. 753, 105 S.Ct. 3420, 87 L.Ed.2d 542 (1985); Confederated Band of Ute Indians v. United States, Northwestern Bands of Shoshone Indians v. United States, or United States v. Minnesota are applicable to this case. In those cases, the documents were found to have clear meanings that were opposed to what the Indians were claiming.
With these principles in mind, we turn to the documents in this case. We have reproduced the entire treaty, 10 Stat. 979, as an appendix to this opinion. The executive orders are not reproduced; they simply set aside certain described tracts of land (or amend the described tracts) as the reservation.
A reading of the 1852 treaty shows that it is a series of promises between the United States and Apache Indians. The Apaches acknowledge the authority of the United States. Both parties promise peace. The Indians promise to treat United States citizens well; the United States promises to *201deal with cases of aggression against the Indians. The Indians promise not to make hostile or predatory incursions into Mexico. Importantly, both Articles 4 and 7 recognize that the Indians have a “territory,” and Article 7 provides that the people of the United States are promised safe passage through it. In Article 9, the United States promises to designate the Apaches’ territorial boundaries at its earliest convenience. In the remainder of the treaty, the United States promises to be liberal and just and to act to secure the prosperity and happiness of the Indians. The treaty expressly states that it is to be given a liberal construction to achieve its ends.
Much historical evidence was presented below. It would unduly lengthen this opinion to recount it all. It will suffice to say that the Tribe’s historians and evidence supported the notion that the reservation’s boundaries were not settled until 1873 because of the intervention of the Civil War, but when they were settled, they were viewed as part of the promises made in the 1852 treaty. Indeed, the foremost expert on Indian law concurs. Felix S. Cohen, Handbook of Federal Indian Law, 302 n. 159 (1971). In contrast, the Water Defense Association’s historians and evidence supported the notion that no sooner was the treaty signed than all parties disregarded it; that no one even knew whether the treaty bound all bands of Apaches, much less all the Mescaleros; that the boundaries were adjusted principally for the convenience of the non-Indian settlers; and that no one contemporaneously recognized any Mescalero or Apache territory, and thus it was proposed to settle the Mescaleros at various times in lands removed from their homelands (those being in the White and Sacramento Mountains, which eventually became their reservation).
It is well established that when the courts below adjudicate matters of fact, we review the issues under a substantial-evidence standard, which accords great deference to the findings made by them. Tapia v. Panhandle Steel Erectors Co., 78 N.M. 86, 89-90, 428 P.2d 625, 628-29 (1967); Mares v. Valencia County Sheriffs Dep't, 106 N.M. 744, 747, 749 P.2d 1123, 1126 (Ct.App.1988). Thus, in this case, we cannot say the trial court was wrong in its findings concerning the historical facts.
Nonetheless, if the question is one of law, we do not accord the same measure of deference. Trigg v. Allemand, 95 N.M. 128, 133, 619 P.2d 573, 578 (Ct.App.1980). Moreover, the question of whether a particular issue is a question of fact or a question of law is itself a legal issue upon which we are free to make our own determination. Edens v. New Mexico Health & Social Servs. Dep’t, 89 N.M. 60, 62, 547 P.2d 65, 67 (1976).
Thus, in this case, we may accept all of the trial court’s findings of fact regarding what actually happened between 1852 and 1873. Those findings, however, do not answer the question of when the reservation was created for purposes of assigning a priority date for its water rights. Nor do the express answers of the various historians when asked the question of when the reservation was created answer this issue. We simply do not believe it is appropriate to treat questions of historical interpretation of dealings between the United States and the Indian tribes as matters of historical fact pursuant to which the substantial-evidence rule exclusively applies.
If, as the Water Defense Association asserts, this means that we are simply superimposing twentieth-century revisionist views onto nineteenth-century history, so be it. Legal requirements sometimes change to reflect the sensibilities of the times. See Schmitz v. Smentowski, 109 N.M. 386, 396, 785 P.2d 726, 736 (1990) (law changes to recognize changing circumstances of evolving society); Stang v. Hertz Corp., 83 N.M. 730, 735, 497 P.2d 732, 737 (1972) (law is dynamic and adaptable to the requirements of society at the time of its application); see also Jones v. Harrisburg Polyclinic Hosp., 496 Pa. 465, 437 A.2d 1134, 1138 (1981) (law must be responsive to new conditions and to sense of justice and social welfare, citing Benjamin N. Cardozo, The Nature of the Judicial Process, 150-51 (1921)). Thus, we ap*202ply twentieth-century notions of fairness and justice to our determination of this issue.
To treat this issue solely as a question of fact would allow a trial court to entirely ignore, as did the trial court in this case, the treaty and the history leading up to it. The rigid approach of divorcing the treaty from the executive orders and analyzing water rights under the restricted purpose analysis proposed by the Water Defense Association and adopted by the trial court both negates the existence of the treaty and fails to recognize the special trust relationship that exists between the United States and the Indian Tribes. Rather, when there is a vague treaty together with a course of dealing that, at best, can be characterized as ambiguous, we believe that the courts must rule, as a matter of law, in accordance with the canons of construction that liberally favor the Indians.
We are supported in this view generally by analogous cases, which we shall describe. We are also supported by language in court opinions that we have discussed above and will advert to below. Most importantly, however, in this case, we are supported by what we believe to be the lynch pin in each case, and that is whether the documents have a clear tenor or an unmistakable meaning. As we have said, they do not, and that is what compels us to rule in accordance with the canons governing liberal construction in favor of the Indians.
Throughout the trial and appeal of this case, the Appellees have pointed out the supposed absurdity of declaring a priority date earlier than the date the boundaries of the reservation were precisely established on the ground that without knowing the boundaries, it is impossible to quantify the federally reserved water. Regardless of difficulties with quantification, which was not attempted for over one hundred years in any event, pertinent cases do not focus on such exacting measures.
Thus, for example, in Walker River Irrigation District, although the reservation was not established by treaty or executive order until 1874, the Interior Department’s setting aside of land for a reservation in 1859 precluded the rights of non-Indian settlers who obtained the same lands in 1860, irrigated them, and claimed prior appropriation. The case was decided on the basis of the liberal construction to be accorded treaties and executive orders due to the dependent status of the Indians.
Similarly, in United States v. Carpenter, 111 U.S. 347, 4 S.Ct. 435, 28 L.Ed. 451 (1884), an 1859 treaty promised that the Indians could use a portion of a certain quarry to make pipes and promised that the United States would survey out the portion necessary to that purpose. Although the quarry was surveyed in 1860, various impediments caused that survey to be insufficient and caused the quarry to have to be resurveyed in 1872. The Court held that a non-Indian who acquired rights to the quarry in 1871 took rights subject to the prior rights of the Indians. This case provides some guidance in answering the State’s specific argument that Congress could not have intended to hang a cloud of uncertainty over great areas of the West at the same time it was encouraging non-Indians to settle there. The same concerns were at issue in Carpenter, and the Supreme Court was not deterred in ruling that the non-Indian settler acquired land subject to Indian rights that were not clearly established at a time prior to the non-Indian’s acquisition of rights.
That the dates of the establishment of precise boundaries are not always controlling is further made apparent by Corrigan v. Brown, 169 F. 477 (W.D. Wash.1907). In that case, the lands were reserved by treaty in 1855, although boundaries were not precisely established until 1873. Nor are we persuaded by language in Minnesota v. Hitchcock, 185 U.S. 373, 390, 22 S.Ct. 650, 657, 46 L.Ed. 954 (1902), which refers to a “certain defined tract.” Hitchcock, like so many other cases we have described, does not state that certainty in the definition of the tract is the sine qua non of a reservation. Rather, the case held that the reservation was established by the defined tract, even though there was not a formal document creating it.
*203We believe that Walker River Irrigation District, Carpenter, and Corrigan support the notion that for purposes of setting a priority date for water rights, the priority date should be the date the United States promised to create a reservation and promised to give that promise a liberal construction, while at the same time exacting promises from the Indians, which subjected them to the authority of the United States. Any contrary holding would be a crabbed interpretation of the dealings between the Indians and the United States, an interpretation that the weight of authority teaches us to avoid. A contrary holding would also be inconsistent with the treaty provisions promising the passage of such laws as may be deemed conducive to the prosperity and happiness of the Indians. Finally, a contrary holding would be inconsistent with the very Winters doctrine upon which the Indians’ water rights are based.
As the remainder of this opinion demonstrates, this rule of liberal construction does not apply to compel courts to rule in favor of the Indians regardless of the type of question presented. Cf. Lopez v. Schultz & Lindsay Constr. Co., 79 N.M. 485, 487, 444 P.2d 996, 998 (Ct.App.) (rule of liberal construction applies to questions of law, not questions of fact) (decided under prior law), cert. denied, 79 N.M. 448, 444 P.2d 775 (1968). Nor, as has already been stated, does it apply to compel courts to rule in favor of the Indians when a clear, contrary result is compelled by the plain language of the treaty or other document.
However, to rule otherwise in a case such as this, which involves a vague treaty and ambiguous contemporaneous dealings, would raise the specter predicted by Justice Douglas in his dissent to Northwestern Bands of Shoshone Indians v. United States, 324 U.S. at 361, 65 S.Ct. at 703: “When the standard [of liberal construction] is not observed, what these Indians did not lose to the railroads and to the land companies they lose in the fine web of legal niceties.” As stated by a more modern commentator, “If one may mark the turn of the 20th century by the massive expropriation of Indian lands, then the turn of the 21st century is the era when the Indian tribes risk the same fate for their water resources.” Joseph R. Membrino, Indian Reserved Water Rights, Federalism, and the Trust Responsibility, 27 Land & Water L.Rev. 1, 14 (1992) (footnote omitted). We believe it is imperative for us to give due regard to the supremacy of federal law in this area and to sensitively recognize the solemn obligation the United States has toward the Indian Tribes because, as state court judges, we will be reviewed under a standard of “particularized and exacting scrutiny.” Arizona v. San Carlos Apache Tribe, 463 U.S. 545, 571, 103 S.Ct. 3201, 3216, 77 L.Ed.2d 837 (1983). Accordingly, we give the law a liberal interpretation and rule that the priority date for the Mescalero water rights is 1852.
PRACTICABLY IRRIGABLE ACREAGE
The same liberality, however, is neither necessary nor desirable in our review of the trial court’s decision applying the PIA standard. The Arizona v. San Carlos Apache Tribe quote noted immediately above was part of a discussion requiring faithful adherence to federal law by state courts. We know of no comparable federal authority requiring us to view the facts in any particular way. Indeed, as will be seen, authority requiring a particular view of the facts would make no sense in the context of appellate review.
Before discussing the specifics of the trial court’s decision, it is necessary to address two preliminary matters raised by the Tribe. The Tribe first complains that the trial court did not make an adequate decision. The Tribe next contends that the trial court erroneously required it to bear the burden of proof.
1. Trial Court’s Decision
The adequacy of the trial court’s decision is challenged because, according to the Tribe, it is neither long enough nor detailed enough in comparison to similar decisions in other cases and considering the wealth of factual material presented below. We disagree.
*204The trial court’s decision is over twelve pages long, consisting of forty-five findings of- ultimate fact and thirty-eight conclusions of law. In comparison, the Tribe points to the special masters’ reports in Arizona v. California, 460 U.S. 605, 103 S.Ct. 1382, 75 L.Ed.2d 318 (1983) [hereinafter Arizona v. California II], Arizona v. California I, and In re the General Adjudication of All Rights to Use Water in the Big Horn River System, 753 P.2d 76 (Wyo.1988), cert. denied sub nom. Shoshone Tribe v. Wyoming, 492 U.S. 926, 109 S.Ct. 3265, 106 L.Ed.2d 610, cert. granted in part sub nom. Wyoming v. United States, 488 U.S. 1040, 109 S.Ct. 863, 102 L.Ed.2d 987, and affd without opinion by equally divided Court sub nom. Wyoming v. United States, 492 U.S. 406, 109 S.Ct. 2994, 106 L.Ed.2d 342 (1989) [hereinafter Big Horn I], each of which ran into hundreds of pages. The Tribe also points to the 900 pages of requested findings, conclusions, and briefs filed by the parties below, including its own requested findings, which numbered in excess of 2000.
We believe the Tribe has been unwarrantedly critical by introducing its briefs on appeal with its complaint about the brevity of the trial court’s decision. First, the trial court here was not a special master and, accordingly, was not required to abide by the detailed reporting requirements of SCRA 1986, 1-053 (Repl.1992). Second, in New Mexico, there is a wealth of law pursuant to SCRA 1986, 1-052(B) (Repl.1992), which provides that a trial court is to find only ultimate facts as opposed to evidentiary facts. Examples of such ultimate facts may be found in Marcus v. Cortese, 98 N.M. 414, 649 P.2d 482 (Ct.App.1982), and the cases cited therein. Upon receiving the parties’ detailed requests for evidentiary findings, the trial court wrote to the parties, commenting on how helpful the requests had been in assisting him to review the voluminous evidence, but asking that the parties comply with the rule and request ultimate findings. That having been done, the trial court then made its decision, properly complying with our rule and our cases.
2. Burden of Proof
The Tribe relies on 25 U.S.C. Section 194 (1982) in arguing that the trial court erroneously required it to bear the burden of proof in this case. Section 194 states:
In all trials about the right of property in which an Indian may be a party on one side, and a white person on the other, the burden of proof shall rest upon the white person, whenever the Indian shall make out a presumption of title in himself from the fact of previous possession or ownership.
The Water Defense Association argues that this statute does not apply to this case for a variety of reasons.
To put the burden-of-proof dispute in context, the issue to which the parties argue the statute either does or does not apply is the PIA issue. Stated otherwise, the dispute is whether the Tribe should bear the burden of production and persuasion to show how many acres it can practicably irrigate or whether the Water Defense Association should bear the burden of production and persuasion to show how many acres the Tribe cannot practicably irrigate. In cases involving PIA, the nature of the proof centers on irrigation projects proposed by the Tribes. See Arizona v. California I & II; Big Horn I. The nature of the proof in this case was no different.
We need not cover all of the parties’ arguments on this issue. On the burden of production issue, to ask the question of who bears the burden is to answer it. Without knowing what projects the Tribe proposes, it is practically impossible for the Water Defense Association to even begin its case. Also, as a logical matter, the Tribe cannot show “previous possession” of future water rights at all, and the only way it would be able to show a presumption of ownership of a certain amount of future water rights would be to go forward with evidence showing feasible irrigation projects.
As to the burden of persuasion, we believe that the same rationale would apply in general. Indeed, in past cases of this na*205ture, the burden of proof has been on the United States. Big Horn I, 753 P.2d at 90; Arizona v. California II, 460 U.S. at 637, 103 S.Ct. at 1400-01. To be sure, 25 U.S.C. Section 194 was not expressly addressed in those cases.
However, in this case, we need not decide whether Section 194 puts the burden on non-Indians in all water adjudications involving PIA. In New Mexico, water rights are adjudicated in two phases, the first phase adjudicating the rights as between the claimants and the State and the second phase adjudicating the rights as between the claimants inter sese. See State ex rel. Reynolds v. Pecos Valley Artesian Conservancy Dist., 99 N.M. 699, 700, 663 P.2d 358, 359 (1983). The two phases took place here during hearings conducted in 1986 and 1987 respectively.
The Tribe does not contend that Section 194 applies to the 1986 hearing. See Wilson v. Omaha Indian Tribe, 442 U.S. 653, 668, 99 S.Ct. 2529, 2538, 61 L.Ed.2d 153 (1979) (Section 194 does not apply to suits between states and tribes). It does contend that Section 194 applies to the 1987 hearing. However, contrary to usual procedure, no subfile orders were entered after the 1986 hearing, and the trial court simply decided the whole case at the end of the 1987 hearing. In arguing about the trial court’s application of the PIA analysis, the parties cite indiscriminately to evidence introduced at both hearings. Under these circumstances, we cannot say that the trial court erred in placing the burden of persuasion on the Tribe, if indeed it did. In this connection, we note that most of the findings do not mention burden of proof at all. Thus, as to most issues, it appears the trial court decided the case not because the Tribe failed in its burden, but rather because the court affirmatively found that the proposed projects were not feasible.
3. PIA Analysis
The Tribe contends that the trial court’s findings and conclusions that the proposed projects do not meet the PIA test are erroneous. The Tribe raises this issue as a legal issue and makes various arguments about why the trial court erred. The Water Defense Association responds that the issue is one of substantial evidence, and that substantial evidence supported all the findings, which in turn supported the conclusions. We agree with the Water Defense Association. In particular, we find nothing to which the Tribe can point that indicates any legal error in the PIA aspects of this case and, as was conceded during oral argument, if the trial court had expressly found that this case turned on the question of the credibility of the experts, there would be no appealable issues in the trial court’s application of the PIA analysis.
In order to understand the parties’ arguments and our decision, it is necessary to outline in some detail the facts of the case. The Tribe proposed two irrigation projects. In the area of the reservation called Rinconada, it proposed to grow barley, potatoes, Christmas trees, cane fruit, strawberries, and apples; in the area of the reservation called Parajita, it proposed to grow alfalfa, barley, corn, potatoes, carrots, Christmas trees, and asparagus. The alfalfa, barley, corn, and potatoes 'are considered field crops; the remainder are considered specialty crops.
To carry out the projects, vast sums of money would have to be invested to build roads, grade fields, and construct the irrigation systems. The irrigation system for the Rinconada project is primarily a series of tunnels and pipelines used to divert water from the Rio Ruidoso and other streams across the mountain to the Rinconada area; the irrigation system for the Parajita project is a series of wells and pumps in the same area as the project.
As to each of the projects and each of the crops within the projects, expert hydrologists, geologists, agronomists, economists, and others testified or submitted reports on soil type and quality, climate and growing season, water quantity and quality, market factors and prices, equipment, labor, and financing. Sometimes, the experts would listen to the testimony of another expert and then critically examine it in their own testimony. Some experts *206were recalled as witnesses as many as half a dozen times.
The definition of PIA used in this case is the same as that used in the Big Horn I case:
“those acres susceptible to sustained irrigation at reasonable costs.” The determination of practicably irrigable acreage involves a two-part analysis, i.e., the PIA must be susceptible of sustained irrigation (not only proof of the arability but also of the engineering feasibility of irrigating the land) and irrigable “at reasonable cost.”
Big Horn I, 753 P.2d at 101. While there is a dispute as to whether certain items in the first prong of the analysis were litigated or stipulated, we need not concern ourselves with arability and engineering feasibility at this time. For purposes of appeal, this case turns on whether the trial court erred in its essential conclusion that the acres were not irrigable at reasonable cost.
The trial court refused a number of the parties’ requested findings on what appeared to be undisputed evidence establishing certain factors of the first part of the analysis. However, this refusal is not error because a trial court is not required to make findings on matters that would not affect the ultimate disposition of the case. See Grants State Bank v. Pouges, 84 N.M. 340, 341, 503 P.2d 320, 321 (1972). The trial court also may have made some findings that appear erroneous. However, this is not error either because erroneous findings of fact unnecessary to support the judgment are not grounds for reversal. See Specter v. Specter, 85 N.M. 112, 114, 509 P.2d 879, 881 (1973). We therefore focus only on the findings that are necessary to support the judgment.
The trial court found that under generally accepted standards for economic feasibility analyses, the projects are infeasible. While the trial court believed that costs incurred off the reservation should be included in the analysis, it expressly found that even if those costs were not included, the projects would not return sufficient funds to pay for their construction, maintenance, replacement, and operation costs. The specific reasons for this finding included: (1) the Tribe’s reliance on specialty crops did not comport with appropriate economic procedures, which consider the proper ratio of specialty crops to basic crops; (2) the Tribe’s analysis of markets for these specialty crops was faulty; (3) the Tribe’s estimates of crop yields were overstated and unrealistic; (4) the terrain and location of the reservation dictated high-quality, top-level management for which the Tribe failed to adequately budget; (5) the Tribe failed to adequately address risks such as weather, insects, and disease; (6) the Tribe failed to include factors such as storage, transportation, supply and demand, and market structure in its budgets; (7) the Tribe understated its labor costs; and (8) the Tribe’s accounting system was inadequate.
Because this is the Tribe’s appeal, the burden is on it to clearly show us how the trial court erred. See Morris v. Merchant, 77 N.M. 411, 416, 423 P.2d 606, 609 (1967); Clayton v. Trotter, 110 N.M. 369, 371, 796 P.2d 262, 264 (Ct.App.1990). In the context of this case, under the above-mentioned cases, this meant that the Tribe had to show, with reference to the best evidence supporting the trial court’s decision, why each finding was error and why any finding that was not error was insufficient to support the judgment. See also Maloof v. San Juan County Valuation Protests Bd., 114 N.M. 755, 759, 845 P.2d 849, 853 (Ct.App.1992) (evidence must be reviewed in light most favorable to judgment and such evidence must be recited in the brief in chief).
The Tribe did not do that. A review of one of the findings, as an example, will show what the Tribe did and why it was insufficient. The trial court found that crop yields were overstated. The Tribe’s experts were predicting crop yields on the reservation, at an elevation of over 6000 feet, that were greater for strawberries than strawberry crops yielded in Michigan, that were greater for potatoes than potato crops yielded in the valleys of Southern Colorado, and that were greater for *207asparagus than asparagus crops yielded in California. To be sure, the Tribe’s experts so testified, basing their testimony on the most modern, scientific farming techniques. However, there was contrary expert testimony indicating that, if such yields were unattainable on farms with better growing conditions, it was unlikely that such yields would be attainable here.
The opinions of experts, even when uncontradicted, are not conclusive on facts in issue, and the trial court may reject expert opinion as it chooses. Sanchez v. Molycorp, Inc., 103 N.M. 148,153, 703 P.2d 925, 930 (Ct.App.1985). In particular, the expert opinions of economists, based as they are on projections, assumptions, and uncertainties, may be rejected by the fact-finder. Strickland v. Roosevelt County Rural Elec. Coop., 99 N.M. 335, 341-43, 657 P.2d 1184, 1190-92 (Ct.App.1982), cert. denied, 99 N.M. 358, 658 P.2d 433, and cert. denied, 463 U.S. 1209, 103 S.Ct. 3540, 77 L.Ed.2d 1390 (1983). Thus, the trial court could reject the Tribe’s expert testimony on crop yields when it appears to border on the fantastic.
The testimony on the other findings was similar to the testimony on crop yields, with the Tribe’s experts testifying to the opposite of what the Water Defense Association’s experts testified to. The Tribe has made only a general acknowledgment of this fact, without relating it to alleged error in specific findings or how those specific findings would be insufficient to support the judgment. There are, then, two reasons why the judgment should be upheld. First, under ordinary substantial evidence principles, when there is testimony going both ways, an appellate court will not say that the trial court erred in finding on one side of the issue. See State v. Chavez, 101 N.M. 136, 138, 679 P.2d 804, 806 (1984) (applying substantial evidence standard to granting of new trial); see also Bagwell v. Shady Grove Truck Stop, 104 N.M. 14, 17, 715 P.2d 462, 465 (Ct.App.1986) (question is whether trial court’s result is supported by substantial evidence, not whether evidence would have supported a different result). Second, the Tribe has not done an adequate job of briefing this issue in that it has not clearly demonstrated that the findings on which there was substantial evidence were insufficient to support the judgment. Rather, the Tribe asserts that there is no way of knowing whether the trial court relied on erroneous principles or satisfactory principles in reaching its result. This assertion stands appellate review on its head. It is not the trial court’s or appellee’s burden to show why a decision is right; it is the appellant’s burden to show why a decision is wrong. Morris v. Merchant.
Apparently realizing this, the approach taken by the Tribe is to attempt to create legal issues out of what appears to be simply matters of evidence. There are three legal issues presented in this way. First, the Tribe claims that in all of its findings, the trial court relied on the Economic and Environmental Principles and Guidelines for Water and Related Land Resources Implementation Studies (March 10, 1983), adopted by the Water Resources Council pursuant to 42 U.S.C. Section 1962a-2 (1982) and approved by the President of the United States and James G. Watt as Chair of the Water Resources Council in 1983 [hereinafter Principles and Guidelines]. Second, the Tribe claims that the trial court erred in applying a perspective that considered benefits or costs outside the boundaries of the reservation instead of applying a so-called reservation perspective to its analysis. Third, the Tribe claims that the trial court erred in failing to make an express finding of the. discount rate it used in its economic feasibility analysis. We discuss each of these items in order.
As to the use of the Principles and Guidelines, these are promulgated to ensure consistent planning by federal agencies, and they apply by their terms to projects by the Corps of Engineers, the Bureau of Reclamation, the Tennessee Valley Authority, and the Soil Conservation Service. They apply a. national perspective so that a project will not be built in Oregon, for example, that will hurt residents of Georgia. They are also fairly conservative, *208so conservative, in fact, that evidence was presented showing that no project has ever been approved under these guidelines.
The Tribe’s issue concerning the Principles and Guidelines is that the trial court erred in following them. There is no indication in the trial court’s findings and conclusions that the trial court considered itself bound by them as a matter of law. To the extent that the trial court relied on them at all, there was testimony, even by the Tribe’s experts, that parts of the Principles and Guidelines provided a useful framework for analysis, and it appears to us that parts of them simply make good sense.
For example, the Principles and Guidelines provide that projected crop yields should be similar to those enjoyed on average in the county. On the other hand, the Tribe contends that the trial court should have adopted its projected yields. Regardless of what the Principles and Guidelines say, it seems to us to make good sense to approach the question of projected yields from the perspective of actual yields, rather than from fanciful projections.
Certainly, a fact-finder hearing evidence on both sides would be entitled to find in accordance with either of the projections. That is consistent with Bagwell. It also serves to explain why the trial court in this case was not required to accept the higher projections of the Tribe, even though the trial court’s counterparts in Arizona v. California and Big Horn I did apparently accept the higher projections.
Similarly, the trial court was not required to find that the Tribe was entitled to rely on specialty crops to the extent it did. It is true that the Principles and Guidelines contain standards on ratio of basic to specialty crops. It is also true that some witnesses testified that part of the reason behind these standards is the national perspective of the Principles and Guidelines, i.e., e.g., they do not want strawberry production in New Mexico to displace strawberry production in Michigan. However, there are in addition sound economic reasons to discourage overproduction of specialty crops. Overproduction would result in a lowering of the market price, thereby lessening the economic' feasibility of a project.
Thus, when the trial court found that parts of the Tribe’s analysis did not comport with appropriate economic procedures, it was not necessarily blindly following inappropriate political guidelines. It was thoughtfully evaluating the evidence before it from a perspective of what is sensible, logical, and in accordance with experience. For these reasons, we cannot say that the trial court erred in relation to the Principles and Guidelines.
The second item is the perspective. The Tribe contends that a fact-finder hearing a water rights adjudication involving Indian claims must use a reservation perspective. By this, the Tribe means that all the evidence should be viewed as the Indians would view it, i.e., if there is unemployment on the reservation such that a project would put people to work, labor costs may be calculated at a zero value; if the project benefits the reservation in other ways, those benefits must be calculated into the analysis; costs off the reservation must never be included in the analysis. We need not decide in this case whether one or another perspective must be used as a matter of law. As a matter of fact, the trial court found that the proposed projects were not feasible even not considering off-reservation costs and “[e]ven making all favorable assumptions.” The Tribe’s argument that this finding cannot be taken seriously is without merit. The Tribe argues that this finding must be disregarded because accepting it would mean that the Tribe should win: if all assumptions in favor of the Tribe were made, then the trial court would have been bound to find in its favor. The Tribe’s argument, however, is contrary to the settled rule that findings are construed in favor of the judgment and not construed to overturn it. See Sheraden v. Black, 107 N.M. 76, 80, 752 P.2d 791, 795 (Ct.App.1988). Accordingly, we construe the finding to mean that even using a reservation perspective for the items required, e.g., deleting 75% of labor costs, as the Tribe’s expert did, and eliminating consideration of *209off-reservation costs, the projects were still not economically feasible. We find nothing in the Tribe’s briefs clearly showing this finding to be wrong.
The third item is the discount rate. The experts agreed that in the abstract, the selection of a discount rate could make or break an economic feasibility analysis. The Principles and Guidelines were using an 8%% discount rate at the time of trial. The Water Defense Association’s experts used a discount rate slightly under that amount. The Tribe’s experts used a discount rate well under that amount, i.e., in the 2% to 4% range. There was testimony that even a zero discount rate could be used to allow for what were called cultural or ethical factors. Essentially, these factors were explained as follows. The point of a discount rate is to discount future benefits to present value so that an accurate determination can be made of whether a project is feasible now. From an ethical perspective, it could be that the future benefits of an agricultural project to an Indian community, which is suffering the economic burdens of generations of depredation and neglect at the hands of the dominant culture, are so great now that no discount rate whatsoever should be applied to these benefits. It was for these reasons that the Tribe’s experts used what appears, from a financial viewpoint, to be an unrealistically low discount rate.
The Tribe contends that the trial court erred in not making a specific finding of fact on what discount rate is appropriate in this case and on what discount rate the trial court chose to use. Because the trial court is required to find only ultimate facts, See Marcus, 98 N.M. at 415, 649 P.2d at 483, we believe the trial court in this case was no more required to find the discount rate than trial courts generally are required to specify the negligent acts in a tort case. See id. Indeed, most civil cases involve future damages that must be discounted to present value. According to the rationale in Marcus, the specific discount rate used in such cases would be an evidentiary fact upon which a specific, separate finding is unnecessary. We do not believe this case is any different.
Nor can we say that the trial court erred in not finding the projects to be feasible and in thereby not finding in accordance with the evidence that was to the effect that a low enough discount rate would make any project feasible. First, there was testimony by the Water Defense Association expert that the projects were infeasible even at a 4% rate. Second, as outlined above, the trial court made findings on the unrealistic nature of the Tribe’s projections on enough of the various factors of the projects that we cannot say, as a matter of law, that adoption of the Tribe’s discount rate would have put the projects over the top of feasibility. The Tribe simply has not shown by reference to the evidence in the light most favorable to the decision that there were insufficient findings supported by substantial evidence to support the judgment.
To sum up on this issue, it appears to us that the Tribe has accused the trial court of considering the variables in such a way that there would never be any practicably irrigable acreage on any Indian reservation. In contrast, as the Water Defense Association argues:
If you use a small enough discount rate, grow an expensive enough specialty crop, assume that market demand will expand, and ignore enough management and labor costs, the standard of economic feasibility adopted by the United States Supreme Court in Arizona v. California becomes meaningless. The word “practicably” has been edited out.
We do not believe that finders of the facts on PIA issues are required by law to adopt any particular view of the facts. In other words, they are neither required to find that specialty crops cannot be grown nor required to find that they can be grown. Their findings are to be based on the evidence before them, and they are to be made with the same sound judgment and good common sense that our legal system expects in all of its cases. Cf. Lopez, 79 N.M. at 487, 444 P.2d at 998 (substantial evidence standard applied to review of *210facts in workers’ compensation case although workers’ compensation act is liberally construed to benefit injured workers, because liberal construction applies to law and not facts).
It may be that in future cases there will be enough variables upon which there is a stipulation or upon which there is undisputed evidence that a particular factor may be viewed to boil down to a question of law. Factors such as how much labor costs should be ignored or what discount rate to apply come to mind. However, the current state of the law does not require a particular bias on any particular factor. In fact, the Special Master’s Report in Arizona v. California II specifically rejected as “misguided” the incorporation of the special subsidies granted to the tribes so that the analysis would be one from the financial point of view of the Indians, Report at 95-96, and rejected as matters of fact some of the tribes’ reliance on specialty crops, Report at 203 (rejecting pistachios); 221 (rejecting figs). Thus, the Tribe’s argument in this case that a particular bias is required appears to be without basis in the authorities cited.
Even if the argument did have merit, we would find it not necessary to decide so here. That is because there was so much disparate evidence on so many factors that we cannot isolate one or several upon which we can say that the trial court applied an incorrect legal view. Although the Tribe has attempted to isolate the discount-rate and labor-cost factors, it has not clearly shown that the other findings were unsupported by the evidence and then were insufficient by themselves to support the judgment.
CONCLUSION
We have considered the parties’ other arguments and find them to be without merit. Accordingly, the judgment is affirmed in part and reversed in part. The judgment limiting the Tribe to a diversion right of 2322.4 acre feet of water is affirmed, but with a priority date of 1852.
IT IS SO ORDERED.
FLORES, J., concurs.
DONNELLY, J., concurs in part and dissents in part.