Tarrant Regional Water District v. Gragg, 43 S.W.3d 609

Tarrant Regional Water District v. Gragg

43 S.W.3d 609

Tex. App.

2001

Check Treatment

OPINION

VANCE, Justice.

Our opinion and judgment dated March 14, 2001 are withdrawn, and the opinion and judgment dated March 21, 2001, are substituted therefor. The only modification is footnote 14 of the Opinion.

The Tarrant Regional Water District (“District”) appeals from a judgment in an inverse condemnation proceeding. The District completed construction of the Richland Chambers Reservoir (“Reservoir”) in 1987. The Reservoir filled by 1989, and began operation. Appellees claim that, due to the presence and operation of the Reservoir, their ranch land was repeatedly inundated with water to the point of being unuseable for ranching. They sued the District for inverse condemnation, and obtained a verdict in excess of $10,000,000. On appeal, the District brings multiple points of error. We will affirm the judgment.

FACTUAL AND PROCEDURAL BACKGROUND

In 1949, O.L. Gragg purchased a ranch of more than 12,000 acres, partly in Anderson County and partly in Freestone County, divided by the Trinity River. Seventeen hundred or so acres are “hill land” which has never been subject to flooding by the river. The remainder is “bottomland” in the Trinity River flood plain. Appellees are the various owners of interests in that ranch. They are grouped in two ways by ownership: fee simple owners and leasehold owners.

In 1987, the District completed the Reservoir, an impoundment of approximately 1.2 million acre-feet of water. It is about eight river-miles north of the northern boundary of the ranch. It is one of two reservoirs operated by the District in the Trinity River watershed, the other being Cedar Creek Reservoir (Cedar Creek), some 35 river-miles upstream from the ranch and about 28 miles upstream from *613 Richland Chambers Reservoir. 1

On May 1, 1991, Appellees filed suit against the District, 2 claiming the District had, by “its intentional and lawful actions, imposed unreasonable restrictions on [Ap-pellees’] use or enjoyment of their land, and interfered with [their] access to their property by virtue of the increased flooding on [Appellees’] property, which is faster, more voluminous, and longer lasting, .... ” Appellees further claimed that the District’s acts “have resulted in a permanent injury to their property interests and a diminution in value of [their] interests in their real property under Article I Sec. 17 of the Constitution of the State of Texas,....” 3 The fee-simple owners claimed damages of $7.15 million and the leasehold owners claimed damages in excess of $4.2 million.

The District’s answer 4 asserted that it had caused no damage to Appellees’ land, that the river had historically flooded the ranch, and that the “temporary and sporadic flooding” resulting from rain did not cause permanent damage to Appellees’ land and thus, no decrease in its fair market value.

The case was tried before a jury in 1998. The court ruled that, as a matter of law, an inverse condemnation had occurred and established the “date of taking” as March 7, 1990. The court made fifty-eight findings of fact and twenty-eight conclusions of law. 5 Damages were determined by the jury. Judgment was entered for Appel-lees, 6 and the District was awarded a “permanent and perpetual flowage easement” over the entire ranch. Obviously, the District appealed.

The District asserts five issues:

1. The evidence pertains only to possible negligent acts and omissions by . the District, not to an intentional “taking,” and therefore there was no inverse condemnation.

2. Appellees failed to present “any evidence” that the District caused the flooding.

*614 3. The trial court erred in failing to conduct separate trials of the “taking” issue and the damages issues, which prejudiced the District throughout the trial.

4. The injuries to the land were temporary and not a permanent “taking,” and the measure of damages was cost of repair and not reduced market value.

5. The jury’s damage awards cannot be sustained because there is “no evidence”: (a) of the difference in the market value of the land with and without the easement, (b) separately proving the value of the bottom lands, la, the flooded lands, and (c) apportioning the damages caused by factors other than the District.

As we view the District’s issues, numbers one, two, and four address the court’s ruling that there was a “taking.” Number five attacks the damages findings of the jury. And number three claims the jury should not have been present at that part of the trial in which evidence of the “taking” was presented. We will first address issue two, then issues one and four together, followed by issue five, and finally issue three.

INVERSE CONDEMNATION

Appellees’ sole claim was that the District had caused an inverse condemnation of their land by its construction and operation of the Reservoir. The applicable elements and standard of review for such a claim are well settled.

The District

The District, as a water control and improvement district, is a political subdivision created under article XVI, section 59 of the Texas Constitution. Tex. Water Code Ann. § 51.011 (Vernon 2000); Bennett v. Tarrant County Water Control and Improvement District Number One, 894 S.W.2d 441 (Tex.App.—Fort Worth 1995, writ denied). It serves only governmental functions. Bennett, 894 S.W.2d at 450. One such function of the District is to provide for the control, storage, preservation, distribution, conservation, and reclamation of water, including flood water. Tex. Water Code Ann. § 51.121(b)(1), (8) (Vernon 2000). It may also control, abate, or change any shortage or harmful excess of water. Id. § 51.121(b)(5) (Vernon 2000). The District is also given authority to acquire easements considered necessary, incident, or helpful to accomplish its purpose. Id. §§ 51.122, 51.123(b), (c) (Vernon 2000). It is undisputed, however, that it never acquired an easement right in the ranch that would authorize it to flood the property.

Texas Constitution

The shield of sovereign immunity does not preclude recovery under “inverse condemnation.” See State v. Biggar, 848 S.W.2d 291, 294-95 (Tex.App.—Austin 1993), aff'd, 873 S.W.2d 11 (Tex.1994). Article I, section 17 of the Texas Constitution provides in part that no person’s property is to be taken for or applied to public use without adequate compensation being made, unless by the consent of such person. 7 Bennett, 894 S.W.2d at 448; Tex. Const, art. I, § 17. An inverse condemnation occurs when property is taken for public use without process or without proper condemnation proceedings, and the property owner attempts to recover compensation therefor. Allen v. City of Texas City, 775 S.W.2d 863, 864 (Tex.App.—Houston [1st Dist.] 1989, writ denied). To *615 recover under the theory of inverse condemnation, the property owner must establish: 1) an intentional act of a governmental entity; 2) accomplished for a public purpose; 3) that damages or takes property from a private citizen. Domel v. City of Georgetown, 6 S.W.3d 349, 357 (Tex.App.—Austin 1999, no pet.) (citing Steele v. City of Houston, 603 S.W.2d 786, 788-92 (Tex.1980)). 8

STANDARD OP REVIEW

The District and Appellees agree that whether a “taking” has occurred, ie., whether there was an inverse condemnation, is a question of law. Id.; see also DuPuy v. City of Waco, 396 S.W.2d 103, 110 (Tex.1965). Conclusions of law are reviewed de novo as legal questions. See Kirkwood v. City of Corsicana, 871 S.W.2d 544, 546 (Tex.App.—Waco 1994, no writ). Thus, we will review the trial court’s determination that an inverse condemnation occurred de novo, that is, “without deference to the lower court’s conclusion.” See State v. Heal, 917 S.W.2d 6, 9 (Tex.1996).

DID THE DISTRICT CAUSE THE FLOODING?

Before addressing the District’s issues about whether the evidence proved a “taking” or only negligence, we review issue two in the District’s brief: “Even if Plaintiffs had a legitimate taking claim, rather than a negligence claim, they failed to adduce any evidence that the District caused the flooding on their ranch that gave rise to that claim.” This is an attack on the court’s findings of causation supporting its ruling that a “taking” had occurred, because without causation, there is no “taking.” The District asserts that the evidence is legally insufficient to establish that the District caused the flood conditions on Appellees’ property, and therefore the trial court’s findings of fact and conclusions of law are incorrect. 9 Specifically, the District attacks the reliability of the opinions of Appellees’ experts offered to prove causation, asserting the opinions were based on inadequate data and flawed methodology, and also attacks the adequacy of the testimony of Appellees’ lay witnesses to prove causation.

To determine whether the evidence is legally sufficient, we consider the evidence “in the light most favorable to the party in whose favor the verdict has been rendered, and every reasonable inference deducible from the evidence is to be indulged in that party’s favor.” We will find the evidence legally insufficient if: (1) there is a complete absence of evidence for the finding, (2) there is evidence to support the finding, but rules of law or evidence bar the court from giving any weight to the evidence, (3) there is no more than a mere scintilla of evidence to support the finding, or (4) the evidence conclusively establishes the opposite of the finding. Merrell Dow Pharm., Inc. v. Havner, 953 S.W.2d 706, 711 (Tex.1997) (citing Robert W. Calvert, “No Evidence” and “Insufficient Evidence” Points of Error, 38 Tex. L.Rev. 361, 362-63 (I960)). “More than a scintilla of evidence exists when the evidence supporting the finding, as a whole, ‘rises to a level that would enable reasonable and fair-minded people to differ in their conclusions.’ ” Burroughs Wellcome Co. v. Crye, 907 S.W.2d 497, 499 (Tex.1995) *616 (quoting Transportation Ins. Co. v. Moriel, 879 S.W.2d 10, 25 (Tex.1994)).

Reliability of the Expert Testimony

Relying on Havner, the District argues that the testimony of Appellees’ experts about the Reservoir causing the flood conditions is legally insufficient because of the inadequate data the experts used when conducting their analysis, and because the analysis itself was flawed. In Havner, the Supreme Court stated that the legal sufficiency of expert testimony to support a finding can be reviewed by examining the reliability of the expert’s data and methodology using the Robinson factors for the admissibility of that same testimony. Havner, 953 S.W.2d at 713-14; E.I. du Pont de Nemours & Co. v. Robinson, 923 S.W.2d 549 (Tex.1995) (adopting the holding setting forth the standards of the admissibility of expert testimony in Daubert v. Merrell Dow Pharm., Inc., 509 U.S. 579, 590, 113 S.Ct. 2786, 2795, 125 L.Ed.2d 469 (1993)). The Robinson factors for determining the reliability of scientific methodology include, but are not limited to:

(1) the extent to which the theory has been or can be tested;

(2) the extent to which the technique relies upon the subjective interpretation of the expert;

(3) whether the theory has been subjected to peer review and publication;

(4) the technique’s potential rate of error;

(5) whether the underlying theory or technique has been generally accepted as valid by the relevant scientific community; and

(6) the non-judicial uses that have been made of the theory or technique.

Havner, 953 S.W.2d at 714 (citing Robinson, 923 S.W.2d at 557). The Havner court concluded:

If the foundational data underlying opinion testimony are unreliable, an expert will not be permitted to base an opinion on that data because any opinion drawn from that data is likewise unreliable. Further, an expert’s testimony is unreliable even when the underlying data are sound if the expert draws conclusions from that data based on flawed methodology. A flaw in the expert’s reasoning from the data may render reliance on a study unreasonable and render the inferences drawn therefrom dubious. Under that circumstance, the expert’s scientific testimony is unreliable and, legally, no evidence.

Id. at 714.

Appellees relied on two engineers, Dwayne Stubblefield and Gary Martin Pet-tit, who are hydrologists, 10 to prove causation. Applying the Robinson factors, the District makes the following allegations about the unreliability of the experts’ testimony.

1. The experts created a computer modeling technique, the X-FOR program, specifically for this case, which has never been used or tested otherwise, was not known in the scientific community, and had no non-judicial use.

2. The X-FOR program was not used to study actual flooding at the ranch or actual reservoir operations. It also did not attempt to follow the route of the water from the Reservoir to the ranch.

3. The data for the X-FOR studies was either hypothetical, or was unreliable according to the experts’ admissions.

4. The X-FOR study was based on the untrue assumption that personnel at the *617 dam always followed the District’s gate release plan when releasing water.

5. The experts rejected other computer models which have been subjected to peer review and publication.

6. The experts did not supply a true rate of error analysis for the X-FOR study.

7. The experts failed to rule out other causes of the injuries to the ranch. Citing Mitchell Energy Corp. v. Bartlett, 958 S.W.2d 430, 448 (Tex.App.—Fort Worth 1997, pet. denied) (an expert proving causation should analyze and exclude other possible causes). These other causes were: record rainfall during the 1990s; water from the Cedar Creek reservoir; seven other reservoirs in the ranch’s drainage area; eight Corps of Engineers lakes in the area; increased runoff from the Dallas/Fort Worth Me-troplex; and tributaries flowing into the Trinity River.

We find two significant flaws in the District’s argument. First, the X-FOR analysis was not the sole or even primary basis for the hydrologists’ opinions about causation. Stubblefield and Pettit testified over a three-day period — some 500 pages of testimony — about the causes of the exacerbated flooding and resultant damage at the ranch. They examined data from the United States Geological Survey of rainfall and water flow, data from the District’s own records (even though they criticized its accuracy because this data understated, the impact of the Reservoir on the ranch), findings from the District’s computer models, data from gauges at the dam and on the river, historical data, and testimony from eyewitnesses on the ground and in the air. The X-FOR program helped analyze part of this data; actually it only made some mathematical calculations to use in producing some charts, and did not advance a new scientific model. The hydrologists compiled and analyzed all this data to examine the pre-Reservoir effects of rain and flow levels, the effects of the gate releases at the dam during rain on water flow down the river versus water flow without the releases, worsened flooding conditions before and after the Reservoir, surges of water post-Reservoir, and the concentrations of water down the river pre- and post-Reservoir. These effects and conditions were studied in relation to the presence and operation of the Reservoir. The hydrologists concluded that the flooding conditions at the ranch were exacerbated by the Reservoir. The District overemphasizes the importance of the X-FOR analysis in this conclusion. Even without the X-FOR analysis, the record supports the conclusion.

Second, the X-FOR program does not fail the Robinson analysis.

• The hydrologists did not believe traditional computer models, like those used by the District, could accurately analyze the available data about the flooding. So, they created a tailor-made model, which in the field of hydrology is an accepted practice. The model they created used techniques and performed studies in a manner accepted and customary in the industry. The model merely helped with their calculations but did not advance a new or novel hydrological theory.

• The hydrologists did follow the route of the water from the Reservoir to the ranch by using the computer models employed by the District.

• The hydrologists used reliable data from numerous sources, as described above.

• Stubblefield did provide an error rate of 10% which was not disputed by the District at trial.

*618 • There was no credible evidence that the cause of the flood conditions was anything other than mother nature or the Reservoir. The hydrologists ruled out mother nature, and found the reservoir to be the cause.

See generally Gammill v. Jack Williams Chevrolet, Inc., 972 S.W.2d 713 (Tex.1998) (not all experts, in particular engineers, are susceptible to a rigorous Robinson analysis, and their personal experience can substitute for at least some of the Robinson factors).

In conclusion, the court, as the “gatekeeper,” could have determined that the underlying data as well as the methodology used by the hydrologists were adequate under Robinson for their testimony to be reliable. See Judge Harvey Brown, Eight Gates for Expert Witnesses, 36 Hous. L.Rev. 743 (1999).

Adequacy of the Lay Testimony

The District also complains that lay witnesses cannot establish causation. Appellees presented numerous lay witnesses familiar with the ranch before and after the Reservoir began operation who testified about the changes in the flood patterns and the effects on the land. Them testimony, taken together, was that after the Reservoir began operation, floods came more quickly not leaving enough time to move the cattle, the floods were more forceful destroying roads and scarring the land, the floods lasted longer and drained slower, and the floods destroyed the crops used to feed the cattle. These conditions did not exist before the Reservoir. This testimony was specifically directed at Appellees’ allegations that flood conditions after the Reservoir were dramatically altered, thereby injuring the land. Credible lay testimony is relevant to causation. City of Odessa v. Bell, 787 S.W.2d 525, 529 (Tex.App. — El Paso 1990, no writ).

Legal Sufficiency

We need not decide whether the testimony of the experts alone, or for that matter of the lay witnesses alone, is legally sufficient to prove causation. Their combined testimony is much more than a mere scintilla to support the court’s findings of causation. See id. (“the testimony of a number of witnesses who were familiar with the water flow of Monahans Draw and with Appellees’ property, as well as the testimony of Appellees’ expert ... is voluminous and consistent in the descriptions ... and was clearly sufficient to support the jury’s finding of cause-in-fact damage to the property.”). Issue two is overruled. We turn to issues one and four.

WAS THERE A “TAKING” OR ONLY A “TEMPORARY DAMAGING”?

The court ruled as a matter of law that an inverse condemnation had occurred. However, the District asserts that even if there was sufficient evidence of causation, the evidence proved only negligence and not the elements of a “taking.” 11 In addition, the District says the evidence shows that the injuries to the land were temporary, not rising to a “taking,” and the measure of damages is the cost to repair the injuries. 12

With reference to whether the established facts support a “taking,” we will follow the long-standing rule that unchallenged findings of fact occupy the same position and are entitled to the same weight as the verdict of a jury. McGalli-ard v. Kuhlmann, 722 S.W.2d 694, 697 *619 (Tex.1986) (citing Swanson v. Swanson, 148 Tex. 600, 228 S.W.2d 156 (1950)). Except for the challenge to the findings of causation, which we have overruled, the District does not challenge the evidentiary support or accuracy of the trial court’s findings. 13 Thus, when we consider the legal question about whether there was a “taking,” we are bound by the findings of fact that the court made. See id.

The Parties’ Contentions

The District asserts the exact opposite of the court’s legal conclusion: that no taking had resulted, as a matter of law. 14 It argues that, although Appellees purported to forego all of their tort claims, 15 their proof showed only negligence. That proof, it contends, will not support an inverse condemnation claim. In addition, the District argues that any harm to Appellees’ land was temporary, not constituting a “taking,” and therefore damages are limited to the costs of repairing the harm, for example, fixing roads and fences. 16 Although it does not challenge the accuracy of the court’s findings of fact, the District places greatly different significance on them.

Conceding that they assert no claim for negligence or an intentional tort, Appellees counter-argument is that the first and second elements of their inverse condemnation cause of action are not in dispute. They say the District intentionally built and subsequently operated the Reservoir for the public purpose of supplying water to its customers. Thus, they say the sole question is the third element — whether that construction and operation resulted in a “taking” of their land.

The Authorities

Under Texas law, a “taking, damage, or destruction” for inverse condemnation purposes is defined as physical appropriation or invasion of property, or unreasonable interference with a landowner’s right to use and enjoy the property. Westgate, Ltd. u. State, 843 S.W.2d 448, 452 (Tex.1992). “Taking” by flooding is a specific type of inverse condemnation and has been the subject of several decisions.

When the City of Graham sued the Brazos River Authority for an inverse condemnation of a sewage disposal plant after the Possum Kingdom Dam was constructed, the Texas Supreme Court asked: D[id] the operation of the dam and the consequent formation of the lake cause the flooding of respondent’s property? Brazos *620 River Authority v. City of Graham, 163 Tex. 167, 354 S.W.2d 99, 106 (1961). Answering in the affirmative, the Court approved a recovery for a “taking” of the plant. Id. at 106. However, the court also stated that in determining whether a “taking” has occurred, the type of property being subjected to the flooding must be considered. Id. at 130 (on rehearing) (“There obviously is a difference in the frequency of floodings which effectually destroy the utility of a sewage disposal plant and the frequency which would impair the usefulness of farming land to the point where such floodings could be regarded as a ‘taking’ of property.”). The Court held that while there had been a “taking” of the sewage disposal plant, a single flooding of the city’s water treatment plant did not result in a “taking” of it. Id. at 106-08. The Court also noted: “the safer rule to follow is one requiring the plaintiff in actions such as this to establish definitely by evidence above the dignity of conjecture that the damage claimed is the result of a repeated and recurring injury rather than a sporadic one. Until a plaintiff is in position so to establish the repetitious nature of the injury, he should be confined in his demand for damages to those flowing directly from the single injury or flooding.” Id. at 108.

A “taking” occurs when the “inundation after erection of the flood-control structures was greater in extent than it previously had been.” Ansley v. Tarrant County Water Control & Imp. Dist. No. 1, 498 S.W.2d 469, 475 (Tex.Civ.App. — Tyler 1973, writ refd n .r.e.). In Ansley, as here, the land in question was not adjacent to the reservoir, but was located some distance away. In fact, it did not border the river, the Trinity River, into which the district was diverting water. The jury found that “The Ansley property or portions thereof has become subjected to repeated increased flooding as a result of the construction, maintenance and operation of the Cedar Creek reservoir, dam, and spillway and discharge channel.” Id. at 471. However, the jury failed to find any diminution in its market value. Id. In affirming the take-nothing judgment, the Tyler Court noted that the finding of repeated, increased flooding gave no indication of the size of the area flooded or the frequency of the “repetition.” Moreover, it said, in view of the fact that the land had always been subject to repeated overflows prior to the construction of the improvements, it was not clear as to whether the finding meant that the jury found that the flooding occurred more frequently after the construction of the improvements or whether the finding merely meant that there was some increased flooding as measured by the magnitude of water thereon. Nevertheless, the court said the finding was sufficient to constitute a finding that construction of the improvements amounted to at least some interference with the plaintiffs’ property rights, but standing alone, would not be sufficient, as a matter of law, to establish their right to recover damages. See id. at 474. Furthermore, the court said that when the evidence shows that injury to land was caused by not only the conduct of the defendant but also by another agency (the Trinity River), an issue is presented about whether the defendant’s conduct, in fact, caused a decrease in the market value of the land. Id. at 475 (“Ap-pellee was liable for the full amount of damages caused by it, but was not liable for the damages caused by the Trinity River.”). In summary, the court rejected the argument that, because the jury found that the land or a portion thereof had been subjected to repeated increased flooding and because the undisputed testimony showed a decrease in value, Ansley’s property had been damaged as a matter of law. Id. at 476 (“The basic error in this argu *621 ment is that it assumes that property which had been floodlands from times immemorial automatically suffers a decrease in market value as a matter of law, if the conduct of appellee contributed in the slightest degree to additional floodings.”). Thus, it concluded, “The inquiry in every ease is whether the property has, in fact, suffered a decrease in market value by reason of the alleged unlawful interference.” Id. at 477.

In Bennett v. Tarrant County Water Control and Imp. Dist. No. One, 894 S.W.2d 441 (Tex.App. — Fort Worth 1995, writ denied), the Fort Worth Court found no taking by flooding because the lake overflowed the dam only once during each of two years, resulting in combined flooding of plaintiffs’ property of little more than twenty days. Id. at 448. The court noted that under federal law, for a Fifth Amendment taking, the critical element of an inverse condemnation in a flooding ease is that of “inevitable recurring floods.” Singleton v. United States, 6 Cl.Ct. 156, 162 (1984). “[Flooding must be intermittent, frequent, and inevitably recurring to constitute a compensable taking, otherwise it is merely a consequential injury or a tort.” Bennett, 894 S.W.2d at 449. The federal courts have established that “one, two, or even three floodings by themselves do not constitute a taking by inverse condemnation.” Id.

But see Golden Harvest Co., Inc. v. City of Dallas, 942 S.W.2d 682, 684, 689-90 (Tex.App. — Tyler 1997, writ denied) (the City’s release in three successive Springs of an abnormal amount of water from the dam at Lake Ray Hubbard, which caused flooding of and extensive damage to Golden Harvest’s land located downstream from the dam, raised a fact issue about whether a “taking” had occurred).

Based on these cases, we believe that the test for whether there was an inverse condemnation by the District is: did the District intentionally perform certain acts in the exercise of its lawful authority to construct and operate the Reservoir for public use which resulted in the “taking” or damaging of Appellees’ property, and which acts were the proximate cause of the “taking” or damaging of such property? See State v. Hale, 136 Tex. 29, 146 S.W.2d 731, 736 (Tex.1941). Because a portion of Appellees’ property flooded pri- or to construction of the Reservoir, the flooding will be considered to be a “taking” only if “inundation after erection of the [reservoir] was greater in extent than it previously had been.” Ansley, 498 S.W.2d at 475. In addition, we are satisfied that isolated instances of increased flooding do not result in a “taking”; “repeated increased flooding” is required. See id. at 474.

Findings of Fact and Conclusions of Law

With respect to conditions on the ranch prior to construction, the court found that the Trinity River has “always flooded the Gragg Ranch.” The court found:

• heavy rains in the Trinity River watershed in 1989, 1990, and 1991 caused heavy flooding all over the Trinity River watershed and on the Gragg Ranch; and

• the floods on the ranch from 1990-97 correlated almost exactly with heavy rainfall in the Trinity River basin.

The court further found that as a “direct result of the construction and operation of Richland Chambers Reservoir”:

• the flooding on the ranch has been far worse than would have occurred under natural conditions, and the flooding will continue to be far worse in the future;

• the flood waters from the lake have arrived at the ranch quicker and *622 with less warning than would have occurred under natural conditions, and will continue to do so;

• the flood waters have arrived at the ranch with more force and velocity than would have occurred under natural conditions, and will continue to do so;

• the floods have been deeper on the ranch than would have occurred under natural conditions, and the deeper floods will continue;

• the ranch has been flooded for longer periods of time than would have occurred under natural conditions, and will continue to be so.

The court further found:

• the existence of a straightened channel leading directly from the dam site almost to the Trinity River results in concentration of the lake discharge waters and decreased travel time of the waters from the dam site downstream to the ranch, as compared to natural conditions;

• the increased flooding caused by the reservoir’s construction and operation and the physical damages to the ranch caused thereby are ongoing and permanent;

• the increased flooding caused by the reservoir’s construction and operation is neither accidental nor the result of isolated or non-recurring events, rather, the flooding is inevitable, will continue permanently, and is the known result of intentional design and operational decisions that are within the District’s power;

• the additional flooding caused by the construction and operation of Rich-land Chambers Reservoir constitutes an actual physical appropriation or invasion of the ranch and an unreasonable interference with the landowners’ right to use and enjoy their property; and

• the flooding renders the bottomland, portion of the ranch virtually useless, greatly diminishing the value of the ranch.

Based on these findings, among others, the court concluded that “the construction and operation of Richland Chambers Reservoir has resulted in a taking, damaging, and destruction of [Appellees’] property by [the District] under Article I, section 17 of the Constitution of the State of Texas.” The court further concluded that there “has been an inverse condemnation of a permanent and perpetual flowage easement encompassing the right of [the District] to flood, overflow and inundate the Gragg Ranch by virtue of the construction and operation of the Richland Chambers Reservoir.”

Review of the Court’s Legal Conclusion

We note at the outset that the District takes the position that damage to Appel-lees’ land resulting from acts concerning the “operation” of the reservoir would, of necessity, be the result only of negligence for which the District cannot be liable. 17 The court made no finding concerning negligence in the operation of the reservoir. Thus, we construe the court’s findings that “construction and operation” of the reservoir resulted in certain events to mean “construction and operation as designed” *623 resulted in those events. Certainly the District intended to operate the reservoir when it constructed it. Furthermore, the court found the flooding was “the known result of intentional design and operational decisions that are within Defendant’s power.” 18

The court’s findings of fact, by which we are bound, establish that the construction and operation of the reservoir as designed resulted in repeated, increased flooding of the ranch, both in degree and frequency. Thus, we are convinced that the court correctly concluded that the District intentionally performed certain acts in the exercise of its lawful authority to construct and operate the Reservoir for public use which resulted in the taking or damaging of Ap-pellees’ property, and which acts were the proximate cause of the “taking” or damaging of such property. Tex. Const, art. I, § 17; Brazos River Authority, 354 S.W.2d at 108; Hale, 146 S.W.2d at 736; Biggar, 848 S.W.2d at 294-95; Ansley, 498 S.W.2d at 475. Issue one is overruled.

MeasüRE of Damages

Because we uphold the court’s conclusion that there was a “taking,” and not a “temporary damaging,” the District’s additional point about the measure of damages being cost of repair rather than reduction in market value is moot. Issue four is overruled.

WERE THE DAMAGES CORRECTLY DETERMINED?

The District asserts that even if its acts caused an inverse condemnation of Appellees’ land, there is “no evidence” to support the jury’s damages award. Specifically, it says there is “no evidence” (a) of the difference in the market value of the land with and without the easement, (b) separately proving the diminution in value of the bottom lands, i.e., the flooded lands, and (c) apportioning the damages caused by factors other than the District, e.g., natural rainfall and other reservoirs. 19

Again, when we review whether the evidence is legally sufficient to support a jury finding, we consider the evidence “in the light most favorable to the party in whose favor the verdict has been rendered, and every reasonable inference deducible from the evidence is to be indulged in that party’s favor.” Havner, 953 S.W.2d at 711. We will find the evidence legally insufficient if: (1) there is a complete absence of evidence for the finding, (2) there is evidence to support the finding, but rules of law or evidence bar the court from giving any weight to the evidence, (3) there is no more than a mere scintilla of evidence to support the finding, or (4) the evidence conclusively establishes the opposite of the finding. Id. (citing Robert W. Calvert, “No Evidence” and “Insufficient Evidence” Points of Error, 38 Tex. L.Rev. 361, 362-63 (I960)). “More than a scintilla of evidence exists when the evidence supporting the finding, as a whole, ‘rises to a level that would enable reasonable and fair-minded people to differ in their conclusions.’ ” Burroughs Wellcome, 907 S.W.2d at 499.

Legal Sufficiency of the Evidence of the Difference in the Market Value of the Land With and Without the Easement

The court instructed the jury to answer only two questions, both on damages: (1) What was the difference in market value, if any, of the “entire fee simple interest” in the ranch immediately before and after the inverse condemnation, taking into consideration any uses to which the Appellees may subject the land after the taking? and (2) What was the difference in market *624 value, if any, of the leasehold estate immediately before and after the inverse condemnation, taking into consideration any uses to which the Appellees may subject the land after the taking? The jury answered $10,214,122 for the fee simple interest and $4,268,547 for the leasehold interest. 20

Appellees proved their damages by two certified land appraisers, Randy Tarpley and Jaimie Wickliffe. Tarpley testified by stipulation that the Gragg Ranch fee (Question one) depreciated in value from $10,000,000 (“before the lake started releasing water”) to $2,850,000 (“after the flooding”). He testified “the Schwertner Priest lease itself, [had] a difference in value, before and after the flooding of $4 million.” (Question two) 21

Wickliffe testified for two days in Appel-lees’ case in chief and also testified in rebuttal to the District’s case, more than 300 pages of testimony in all. She described in great detail the appraisal techniques and standards she used and the data she analyzed when determining how to compute damages, such as sales of other properties in the area, the number of cattle the ranch could support, and their value. At one point she was asked “... in a condemnation ease like this, will you tell the jury whether or not you’re required to appraise the property before the alleged taking, and its condition before whatever the damaging factor is alleged to be, and in its condition after the damaging factor, in this case, the construction and operation of the lake,” to which she responded: “That’s what — exactly what you have to do. You need to look at the property before in its before condition and analyze it, and then look at the property separately in its after condition.” 22 She also answered “yes” to this question: “In your report, Mrs. Wickliffe, I notice — in summary, Mrs. Wickliffe, are you of the view that your estimate of the damages suffered by the Gragg family, and the diminished value of the lease suffered by the Schwert-ner and Priest families accurately and conservatively represent the lost values as a result of the existence and the operation of Richland-Chambers reservoir?” 23 A chart of her conclusions was eventually admitted as Exhibit 136, which showed the loss in value of the ranch to be $10,214,122, the amount found by the jury in Question one, and the loss in value of the Schwert-ner/Priest lease to be $4,268,547, the amount found by the jury in Question two. 24

Based on this testimony, we find the evidence to be legally sufficient to support the jury’s answers to the two questions.

Legal Sufficiency of the Evidence (1) Separately Proving the Diminution in Value of the Bottom Lands and (2) Apportioning the Damages Caused by Factors Other Than the District

The District complains there was no evidence to support nonexistent jury findings. The jury was not asked about the diminished value of only the bottom lands, or about damages caused by factors other than the District. We cannot review a sufficiency point about a jury finding which does not exist. If the District intended to raise an issue about a charge error, ie., that the trial court put the measure of damages to the jury erroneously, it has failed to do so. See Tex.R.App. P. 38.1(e).

*625 Even if the District had raised charge error on appeal, its complaint about the separate diminution in value of the bottom lands is unpreserved. The District filed written requests for questions and instructions to the charge, none of which specifically addressed this complaint. 25 At the charge conference before closing statements, the District orally made objections and requests to the charge, but again, none specifically addressed this complaint. 26 We cannot review unpreserved complaints. Tex.R.App. P. 33.1.

As to the complaint about other causes of the damages, the District did file written requests, and also orally requested at the charge conference, that the jury be instructed “not to include any amount of money for a difference in value to the [property] unless you find from a preponderance of the evidence that such difference in value ... is not due to some cause other than the Richland Chambers Reser-vón*, such as damages to the Ranch caused by flooding on the Trinity River that would otherwise have occurred, the construction of the bridge over the Trinity River by O.L. Gragg, or the levees/roads constructed on the Gragg Ranch by O.L. Gragg.” These were not granted.

A trial court’s failure to submit a requested instruction in the charge is reviewed for abuse of discretion. See Texas Dept, of Human Services v. E.B., 802 S.W.2d 647, 649 (Tex.1990); Knoll v. Neb-lett, 966 S.W.2d 622, 633 (TexApp.- — Houston [14th Dist.] 1998, no pet.). An abuse of discretion occurs “when the trial court’s decision is arbitrary, unreasonable, and without reference to guiding principles.” Mercedes Benz Credit Corp. v. Rhyne, 925 S.W.2d 664, 666 (Tex.1996). If abuse is found, the judgment will be reversed only if the failure probably caused the rendition of an improper judgment. Tex.R.App. P. 44.1(a)(1); Macias v. Moreno, 30 S.W.3d 25. 27 (Tex.App. — El Paso 2000, no pet.).

The jury was charged immediately before the two damages questions with this instruction: “You are instructed that in answering the following questions, you are to consider only the differences in value caused by the construction and operation of Richland Chambers Reservoir.” In substance, this fulfilled the District’s request. Therefore, we find no abuse of discretion in refusing the requested instruction.

Issue five is overruled.

SHOULD THE ISSUES HAVE BEEN TRIED SEPARATELY?

Whether there was a “taking” is a question of law for the court to decide. State v. Heal, 917 S.W.2d 6, 9 (Tex.1994). However, the amount of damages is a question of fact for the jury. Hams County v. Felts, 881 S.W.2d 866, 870 (Tex. App. — Houston [14th Dist .] 1994), affd, 915 S.W.2d 482, 484 (Tex.1996). The District complains that the trial court, without a jury, should have heard the evidence about whether there was a “taking,” and if the court held there was a “taking,” only then should a jury have heard evidence on and decided damages. 27

The District presents no authority that in Texas the preferred practice is to separate the issues in an inverse condemnation trial. The District cites State v. Wood Oil Distributing, Inc., 751 S.W.2d 863, 865 (Tex.1988), which states “It is incumbent *626 upon the trial court to make [the taking] determination prior to trial and to control the admission of evidence accordingly.” However, in Wood Oil, the trial court excluded evidence of damages for inconvenience due to circuitry of travel in an impairment-of-access “taking” case. The appeals court reversed, holding the jury should have heard this evidence. The Supreme Court reversed the appeals court because those damages are not compensa-ble in an impairment-of-access case. This led to the statement quoted above. The District cites no other case to support its argument. 28

Resolution of the District’s issue is controlled by Tex.R. Civ. P. 174(b), which states: “Separate Trials. The court in furtherance of convenience or to avoid prejudice may order a separate trial of any claim, cross-claim, counterclaim, or third-party claim, or of any separate issue or of any number of claims, cross-claims, counterclaims, third-party claims, or issues.” [Emphasis added]. There was only one claim — an inverse condemnation. So, the separation concerns the damages issue versus the “taking” issue.

The District alleges it was prejudiced by the non-bifurcated trial in six ways: (1) the District was confused throughout the trial about whether the judge or the jury would decide the “taking” question; (2) the District was forced to deny liability to the jury, while at the same time presenting evidence on the value of the damages; (3) the District was confused about whether Appellees claimed a “taking” or a “damaging,” ie., about whether Appellees wanted “taking” damages or only damages to repair; (4) evidence of easements obtained from other property owners through set-, tlement was adduced, which evidence was improper and inflamed the jury; (5) Ap-pellees successfully excluded evidence of injury throughout the Trinity River basin from natural causes, but argued to the jury that the District offered no evidence that damages occurred to property upstream from the Reservoir; and (6) most of Appellee’s closing argument to the jury went to causation, not valuation, “resulting] in a tainted verdict.”

The District further contends “because the evidence and issues are discrete,” separate trials of the “taking” and damages issues would not have been inconvenient. Also, the District argues that judicial economy was not promoted by the single trial because if, in a separate trial, the judge had found no “taking,” there would have been no need for the second part of the trial on damages.

Taking these arguments in reverse order, because the trial court found a “taking,” the effect of a separate trial on judicial economy is hypothetical, and therefore moot.

As for whether a separate trial would not have been inconvenient because the “taking” and damages issues are discrete, the District has the question reversed. In a Rule 174(b) separate trial of issues, the question is whether the action the court took, ie., a single trial, was convenient. See Womack v. Berry, 156 Tex. 44, 291 S.W.2d 677, 683 (Tex.1956). The convenience of a single trial is one ground on which separate trials are rejected. Id.; Rule 174(b). The District’s argument that separate trials would not have been inconvenient does not address whether the single trial was convenient. Therefore, the District presents no reviewable-issue.

*627 Numbers two, four, five, and six of the allegations of prejudice contend that the jury was misled by the single trial, for a variety of reasons. We fail to see how any of the complained-about effects of the single trial may have affected the jury’s determination of damages. As noted, the jury was asked to find:

1. the difference in market value of the entire fee simple interest in the Ranch immediately before and immediately after the inverse condemnation of the easement, and

2. the difference in market value of the Schwertn er/Pri est Leasehold estate immediately before and immediately after the inverse condemnation of the easement.

The jury’s answers to these two questions were derived from testimony from two appraisers about the value of the land before and after the Reservoir. The District has not shown how the District’s having to deny liability while at the same time attacking damages, evidence about other easements upstream, and closing argument by Appellees about causation, may have affected the jury’s answers to the questions.

We find no merit in allegations of prejudice one and three. The claim is the District was confused about the nature of Appellees’ claims and damages, and the law pertaining thereto. The lawsuit was filed in 1991, and trial was in 1998. The parties tried the case on Appellees’ Eighth Amended Original Petition and the District’s Fifth Amended Answer. The record reflects extensive discovery. If the District was in fact confused about the nature of Appellees’ pleadings or about the law applicable thereto, it cannot fault the single trial for that confusion.

Finally, it made sense for the issues not to be separated. The two questions presented to the jury were preceded by this instruction: “You are instructed that in answering the following questions, you are to consider only the differences in value caused by the construction and operation of Richland Chambers Reservoir.” [Emphasis added]. The evidence about causation applied both to the “taking” issue and to the damages issues. The trial court could not have known at the beginning of trial what the evidence would be at the end of the trial, or how much of it the jury should hear to answer whatever questions were to be presented. The trial court .exercised its discretion in avoiding the possibility of needless repetition of the evidence, which would have required many extra days. 29 A trial court’s decision not to separate issues is reviewed for abuse of discretion. Womack, 291 S.W.2d at 683. There is no duty to order a separate trial unless “all of the facts and circumstances of the case unquestionably require a separate trial to prevent manifest injustice, and there is no fact or circumstance supporting or tending to support a contrary conclusion [i.e., not to separate], and the legal rights of the parties will not be prejudiced” by a separate trial. Id. The record does not support a conclusion that the trial court abused its discretion.

Issue three is overruled.

CONCLUSION

Having overruled all the District’s issues, we affirm the judgment.

Notes

. The drainage area of the Trinity River upstream from the ranch is 12,000 square miles and encompasses Dallas and Fort Worth and extends almost to Wichita Falls. There are nine water conservation reservoirs and eight flood control lakes within that portion of the watershed.

. Appellees' trial pleading was their Eighth Amended Original Petition, which the court authorized them to file.

. Appellees specifically waived any claim under the United States Constitution and any claims based on negligence or an intentional tort. See generally Mayhew v. Town of Sunnyvale, 964 S.W.2d 922, 933 (Tex.1998) (citing Williamson Planning Comm’n v. Hamilton Bank, 473 U.S. 172, 175 n. 1, 105 S.Ct. 3108, 3110-11 n. 1, 87 L.Ed.2d 126 (1985)) (Fifth Amendment provision about takings has been incorporated through the Fourteenth Amendment to apply to the states). Prior to the trial on the merits, Appellees’ tort claims were the subject of an interlocutory appeal to this court. We stayed further action on all claims other than the inverse condemnation claim. See Tarrant Regional Water District v. Gragg, 962 S.W.2d 717, 719 (Tex.App.—Waco 1998, no pet.).

. The District's Fifth Amended Answer was, to the extent applicable, allowed as the response to Appellees’ Eighth Amended Petition, even though it was filed earlier.

. At Appellees’ request the court adopted three additional conclusions of law and corrected one conclusion. At the District's request, the court granted twelve additional findings of fact, some with modification. The court also denied fourteen of the District’s requested additional findings of fact, rejected all proposed additional conclusions of law, and overruled all the District’s objections to the Court’s Findings of Fact and Conclusions of Law.

. $4,268,547 to the owners of the leasehold, and $5,945,575 to the owners of the fee estate.

. Because Appellees waived any claim under the United States Constitution, we address only their rights under the Texas Constitution.

. Cf. Waddy v. City of Houston, 834 S.W.2d 97, 102 (Tex.App.—Houston [1st Dist.] 1992, writ denied) (the elements of a taking are: 1) the governmental entity intentionally performed certain acts; 2) that resulted in a taking of property; 3) for public use).

. In footnote 20 of its brief, which we will disregard, the District also attacks the factual sufficiency of the evidence to support unspecified findings of fact.

. Hydrology is "a science dealing with the properties, distribution, and circulation of water....” Merriam Webster's Collegiate Dictionary (10th ed.1993).

. Issue 1, Appellant's Brief.

. Issue 4, Appellant’s brief.

. As we will see, the District’s fifth issue raises "no evidence” questions about the special issues inquiring about the market value of the ranch before and after the "taking.” As we view the issue under consideration, those attacks are not directed against the court’s finding that there was a taking.

. Amicus curiae briefs were filed in support of the District’s position by the Brazos River Authority, Canadian River Municipal Water Authority, City of Dallas, City of Houston, Lower Colorado River Authority, Lavaca Na-vidad River Authority, San Jacinto River Authority, and the Texas Water Conservation Association. Tex.R.App. P. 11. The Texas Farm Bureau filed an amicus curiae brief on behalf of Appellees.

. The District asserts that it enjoys sovereign immunity from tort claims, except to the extent that the Texas Tort Claims Act waives its immunity. Bennett v. Tarrant County Water Control and Improvement District Number One, 894 S.W.2d 441, 450 (Tex.App. — Fort Worth 1995, writ den’d). As we will see, this issue is not before us.

. The District includes in this issue an objection to instructions in the charge about the perpetual flowage easement, claiming the instructions were overly broad because the flooding was "only temporary or sporadic ... tied to rainfall.” This objection is resolved by resolution of the primary issue, i.e., whether there was a "taking."

. Specifically, it says, "Inherently, then, if the reservoir is operated as intended, it would most frequently reduce downstream flow and, in any event, would not add more water downstream than would pass by naturally. Only if the District were negligent in operating the reservoir would it add more water than would have passed downstream naturally. Indeed, the nature of the conduct complained about by [Appellees] is negligence. ..."

. Finding of Fact 22.

. Issue 5, Appellant’s brief.

. The owners of the fee estate are entitled to $10,214,122 less $4,268,547, or $5,945,575.

. RR volume VII, pp. 1-2.

. RR volume XV, pp. 38-39.

. RR volume VIII, p. 92.

. RR volume XV, p. 32.

.CR volume 8, # 1498. In fact, the District's requested questions would have made the same damages inquiries that the court’s charge made.

. RR volume XV, pp. 1-9.

. Issue 3, Appellant's brief.

. See Iley v. Hughes, 158 Tex. 362, 311 S.W.2d 648, 651 (Tex.1958) (liability and damages in personal injury cases should not be separated under Rule 174(b), because “[Ojur courts have always frowned upon piecemeal trials....”).

. The trial lasted fifteen days, during which twenty-four witnesses were called and 250 exhibits were admitted.