*1022 Opinion

McKINSTER, J.

In this matter we are asked to determine whether or not the State of California (hereinafter State or petitioner) owns all of the groundwater present under the surface of the State. We conclude that it does, but that its “ownership” is not necessarily such as to trigger an “owned property” exclusion in a policy of liability insurance. Accordingly, we set aside the trial court’s ruling that the “owned property” exclusion applies, and remand for further proceedings on the question.

This action has its genesis in the State’s maintenance and/or operation of a toxic waste facility commonly known as the Stringfellow Acid Pits. In related litigation, the State has been found liable for massive damages related to environmental cleanup costs. In the hope of covering these enormous expenses, the State has turned to real parties in interest, the issuers of various insurance policies arguably covering the State’s liability. Real parties in interest have balked. In the instant litigation, the State seeks declaratory relief with respect to the insurers’ duties to defend and indemnify it in the related actions, and also seeks damages based on a number of versions of breach of the implied covenant of good faith and fair dealing.

Due to the complexity of the matter, the litigation has been divided into four parts. The first part, or phase I, involved “policy prove-up issues.” That phase is not involved here. In phase II, the court was to address the “legal interpretation of key provisions of the policies, separate from the application of the facts of this claim.” Phase III would consider the application of the facts of this case to the policies, and phase IV, if necessary, would address, inter alia, issues of bad faith and damages. It is phase II with which we are concerned here.

To be more specific, the particular policy provisions in question in this petition are what are customarily referred to as “owned property exclusions.” By means of such exclusions, the insurer notifies the insured that the insurer will not pay for loss or damage to property owned by the insured. 1 By way of example, a policy of automobile liability insurance will commonly contain such an exclusion, which means that if the insured’s own car is damaged, the insurer will not pay. If, in contrast, the owner also procures collision and comprehensive coverage, these coverages will apply to the insured’s own vehicle and no “owned property” exclusion is appropriate.

*1023 In 1998, after hearing, the trial court proposed to issue an order in phase II which would have found that the “owned property” exclusion did not apply because the groundwater, claimed to have been contaminated due to the operation of the Stringfellow Acid Pits, did not belong to the State. However, before the court finalized this tentative ruling, real parties in interest sought leave to present additional argument and authorities on the issue, and a new round of hearings and briefing took place. In short, real parties in interest this time convinced the trial court that the “owned property” exclusion was not ambiguous and did apply. 2 It is this ruling that the State assigns as error. 3

Discussion

We will first consider the statutory and other authorities which tend to lead to a conclusion that the State does own the groundwater. We will then move to those authorities which might produce a contrary result. Finally, we will deal with the effect of recent pronouncements by our Supreme Court that, taken on their face, appear to dispose of the question.

The State Owns the Groundwater

Several statutes, one in the Water Code and two in the Civil Code, support ■ the ruling of the trial court and the argument of real parties in interest. First, Water Code section 102, enacted in 1943, provides that “All water within the State is the property of the people of the State, but the right to the use of water may be acquired by appropriation in the manner provided by law.” Civil Code sections 669 and 670 together provide that “All property has an owner’’ and “[t]he State is the owner . . . of all property of which there is no other owner.” Thus, runs the argument proposed by real parties in interest, the People of the State own the groundwater involved in this case under the Water Code; furthermore, even without this provision, the State must be held to be the owner by default, as individuals cannot own such water.

The latter part of the proposition is certainly correct, as there is no private ownership of ground or flowing water—a subject to which we will return. Furthermore, the language of the Water Code provision appears to be plain and unequivocal.

However, the matter is not as simple as it seems at first blush. The State responds to this apparently conclusive authority by arguing that the Civil *1024 Code concept of ownership does not apply to groundwater and that the Water Code provision quoted above does not establish a proprietary interest, but expresses more of a philosophical view. Authorities of substantial weight support this position.

The State Does Not Own the Groundwater

The first point to make is that modem water law focuses on the concept of water rights rather than water ownership. (1 Waters and Water Rights (1991 ed.) § 4.01, p. 65.) 4 Most notably in the western states, water rights, in turn, are frequently tied to water use. Most commonly, a condition to the right to use 5 is that the use be beneficial. Such an approach was essential for the economic and social development of regions where the natural rainfall was either unreliable or routinely inadequate to support the fertility and profitable agriculture otherwise promised by the climate and the soil. (See Katz v. Walkinshaw (1902) 141 Cal. 116, 124-127 [74 P. 766], for a discussion of the conditions that prompted the reasonable or beneficial use doctrine.)

This state initially recognized both riparian and appropriative rights with respect to both surface water and groundwater. (Lux v. Haggin (1886) 69 Cal. 255 [10 P. 674], passim; Katz v. Walkinshaw, supra, 141 Cal. at p. 150.) Although the Supreme Court held in Herminghaus v. Southern California Edison Co. (1926) 200 Cal. 81, 105 [252 P. 607] that a riparian owner was entitled to the entire flow as against an appropriator, without regard to reasonableness, two years later this was qualified by the predecessor to section 2 of article X of the California Constitution. That section, after setting forth a general policy against waste and in favor of beneficial use, provides that “The right to water or to the use or flow of water in or from any natural stream or water course in this State is and shall be limited to such water as shall be reasonably required for the beneficial use to be served . . .” and even riparian rights are so limited. The general rules of priority— for example, between holders of riparian rights and appropriators—apply equally to surface and groundwaters. (See City of Pasadena v. City of Alhambra (1949) 33 Cal.2d 908, 925-926 [207 P.2d 17].)

*1025 Thus, the current state of the law is that a riparian (or overlying) owner, or an established appropriator, has the right to take and use water from, e.g., a flowing stream, but the flowing water is not owned. On the other hand, a water right itself has been considered an interest in real property. (See, e.g., Schimmel v. Martin (1923) 190 Cal. 429, 432 [213 P. 33].) It is also sometimes described as a right “appurtenant to” or “part and parcel of’ an interest in real property. (See, e.g., Lux v. Haggin, supra, 69 Cal. at pp. 390, 391-392.)

It may be true that, at least prior to the 1928 adoption of the predecessor to section 2 of article X of California Constitution, one could speak of “ownership” of water itself (Lux v. Haggin, supra, 69 Cal. at p. 392), and there obviously remains a sense in which discrete quantities of water can be “owned.” For example, one who purchases a container of Arrowhead Puritas water then “owns” five gallons of California water. (See Lewis v. Scazighini (1933) 130 Cal.App. 722, 724 [20 P.2d 359], recognizing that water severed from the land becomes personal property which may be bought and sold like any other commodity.) But in its natural state, water is certainly not subject to ownership by an individual. 6

But in what sense is it owned by the State?

“Ownership” and “Ownership”

Given the explicit provisions of Water Code section 102, the position that the State has no kind of ownership over the waters is not tenable. However, as petitioner points out, the Legislature did not enact that the State owns the water, but that the water is owned by the People of the State. In our view, although the difference is somewhat nebulous, it confirms the State’s position in general.

In Civil Code section 670, quoted in part above, the Legislature also provided that “The State is the owner of all land below tide water, and below ordinary high-water mark. . . .” (Italics added.) The State is considered to hold “title as trustee to such lands and waterways. . . .” (National Audubon Society v. Superior Court (1983) 33 Cal.3d 419, 434 [189 Cal.Rptr. 346, 658 P.2d 709], italics added.) Thus, the use of the term “The State” in section 670 appropriately conveys the sense that the State, as a governmental entity, owns the tidelands and waterways.

*1026 The term “The People of the State” is evidently designed to express a more abstract notion. 7 The Legislature’s use of the term is entirely consistent with an intent of expressing a regulatory or supervisory power rather than anything even approaching a proprietary interest or the right to exercise physical dominion. Certainly, as is clear from the discussion of water rights above, the statute does not mean that any particular Californian has a right to take and use any particular water, without regard for riparian land ownership or appropriative rights. 8 A California citizen could not rely on Water Code section 102 to legalize his surreptitious taking of water to the detriment of the holder of an adjudicated or common law right to the water. Nor does the statute confer upon the State any such right to take any particular water or mandate any particular flow; when it comes to removing or controlling the natural waters, the State is bound by the same rules applicable to private landowners.

Thus, in Fullerton v. State Water Resources Control Bd. (1979) 90 Cal.App.3d 590 [153 Cal.Rptr. 518], the State Department of Fish and Game sought a permit for an “in-stream” appropriation of water in the Mattole River to maintain water flow for the protection of fish. 9 In a ruling upheld on appeal, the application was denied on the basis that its proposed use did not meet the statutory and common law requirements for an appropriation. No contention was made that the Department of Fish and Game, as an arm of the State, could simply arrange for the increased flow in derogation of the rights of recognized riparian and/or appropriators.

Indeed, early decisions of the courts of this state interpreting statutory and constitutional provisions relating to water repeatedly held that the State could not lawfully—that is, constitutionally—interfere with established water rights. (See, e.g., Lux v. Haggin, supra, 69 Cal. at p. 368.) In City of San Bernardino v. City of Riverside (1921) 186 Cal. 7, 29-30 [198 P. 784], the court scoffed at the notion that the predecessor to Water Code section 102 should be construed as granting proprietary, possessive ownership of all waters in the state to the State. Although it recognized that literally the *1027 statute could be so read, it soundly rejected any such interpretation, commenting that “The water that pertained to or was contained in the lands of the state was already the property of the people when the amendment was adopted. The statute was without effect on any other property.” (186 Cal. at p. 30.) By this language, the court recognized that the State’s proprietary interest in waters derives not from the general pronouncements of the Water Code, but from the same bases that give rise to water rights in individuals as riparians or appropriators.

It is perhaps unfortunate that the word “property” is sometimes used in the cases and statutes with no careful consideration of the nuances of its meaning. 10 The same may be said to be true of the terms “own,” “owner,” and “ownership.” We return now to the Civil Code provisions on which the parties variously rely.

We have cited above Civil Code section 670, which provides that “The State is the owner . . . of all property of which there is no other owner” and is therefore relied upon by real parties in interest. The State, however, points out that “property,” in turn, is defined as “the thing of which there may be ownership” and that “ownership” is “the right of one or more persons to possess and use [a thing] to the exclusion of others.” (Civ. Code, § 654.) A thing may be owned if it is “capable of appropriation or of manual delivery. . . (Civ. Code, § 655.) Real property includes “That which is incidental or appurtenant to land” (Civ. Code, § 658) while everything—that is, all property—that is not real property is personal property. (Civ. Code, § 663.)

We agree with the State that it does not “own” the water of the state in its natural conditions within the meaning of these statutes. 11 As we have explained above, the State does not have the right to “possess and use [the water] to the exclusion of others” and has only such riparian or appropriative rights as it may otherwise obtain by law. It is also to be repeated that riparian water rights have traditionally been held to be a type of real property (Schimmel v. Martin, supra, 190 Cal. 429) and this is consistent with Civil Code section 658 insofar as the statute includes as “real property” anything which is “incidental or appurtenant” to land.

Real parties in interest correctly point out that the Civil Code recognizes both absolute and qualified types of ownership. (Civ. Code, §§ 679, 680.) *1028 “Qualified” ownership means ownership which is either shared with others, in which the time of enjoyment is deferred or limited, or where the use of the property is restricted. (Civ. Code, § 680.) The statute has received no useful judicial construction; in Potts Drug Co. v. Benedict (1909) 156 Cal. 322, 331 [104 P. 432], which involved a leasehold, the court limited its relevant comments to the obvious observation that one may own a thing although another has the right to the present use and possession of it.

However, where a thing is subject to rights which limit the owner’s rights, the quintessential element of ownership is that the owner’s right’s increase as those of the other interested party decrease or are extinguished. (Rest., Property, § 10.) The application of such a principle to the situation in which an owner has granted a leasehold interest to another (e.g., Potts Drug Co. v. Benedict, supra, 156 Cal. 322), or in which property is subject to a life estate, is clear. When the tenancy expires, or the life tenant expires, the rights of the owner of the underlying estate expand. In our case, however, the State’s purported rights of “ownership” as granted by the Water Code do not increase at all or under any circumstances. Even if holders of all the established water rights in the state should cease to exercise them, the State would have no greater power over the water than it had before. And as the court confirmed in City of San Bernardino v. City of Riverside, supra, 186 Cal. 7, the State cannot by fiat simply erase or alter established water rights to its own benefit.

Put another way, the State’s “ownership” powers are not “restricted” by the holders of water rights; the State’s “ownership” under Water Code section 102 confers no powers of possession or use upon it. Read in context with provisions and decisional law relating to water use, the Water Code provisions simply do not result in anything recognizable as “ownership” as the term is commonly understood.

The Effect of AIU Ins. Co. v. Superior Court

Real parties in interest, and the trial court, rely heavily on language found in AIU Ins. Co. v. Superior Court, supra, 51 Cal.3d 807, in which a policyholder somewhat similarly situated to the State here sought coverage for toxic cleanup costs from its insurers. First, in a footnote, the Supreme Court commented that “The third party suits allege not only contamination of the disposal sites themselves, but also of groundwater, surface water, and aquifers on and surrounding the sites. [The plaintiff] cannot be considered the owner of such water. (See, e.g., Wat. Code, § 102 . . . .)” (51 Cal.3d at p. 817, fn. 6.) Later in the opinion, discussing whether governmental “response *1029 costs” were “damages” under the plaintiff’s policy, 12 the court stated that “release of hazardous waste into groundwater and surface water constitutes actual harm to property in which the state and federal governments have an ownership interest. . . (Id. at p. 829, italics added.)

As a lower court, we are bound by decisive pronouncements of the Supreme Court. (Auto Equity Sales, Inc. v. Superior Court (1962) 57 Cal.2d 450, 455 [20 Cal.Rptr. 321, 369 P.2d 937].) Conversely, we are not bound by dicta. (Fireman’s Fund Ins. Co. v. Maryland Casualty Co. (1998) 65 Cal.App.4th 1279, 1301 [77 Cal.Rptr.2d 296].) 13 However, disagreement with clear language in a Supreme Court opinion concededly puts a lower court in a “delicate position.” (Ibid.)

We are not as sure as the State that the crucial comments in AIU Ins. Co. are dicta 14 but we do agree that the Supreme Court cannot, in six lines of a case devoted to insurance law, have intended to create a new conceptual scheme relating to water law. The high court’s sole supporting reference is to Water Code section 102 and its comments must be read in that context. That is, AIU Ins. Co. stands for no more than that the state has an “ownership” interest as expressed in the Water Code. The court devoted no consideration or analysis with respect to the statute and that task has fallen to us. 15

*1030 The Nature of the State’s Ownership

In our view, Water Code section 102 is an example of what the United States Supreme Court has called a “ ‘ “fiction expressive in legal shorthand of the importance to its people that a State have power to preserve and regulate the exploitation of an important resource.” ’ . . .” (Sporhase v. Nebraska ex rel Douglas (1982) 458 U.S. 941, 951 [102 S.Ct. 3456, 3461, 73 L.Ed.2d 1254], citations omitted.) This power, of course, derives from the police power conferred by the United States Constitution. (Id. at p. 956 [102 S.Ct. at p. 3464].) Water Code section 102 thus expresses the preeminent right of the people of the State to make water policy and control water usage; it may perhaps also have been intended as a preemptive strike against any private effort to claim “ownership” in a proprietary sense. But the State’s power under the Water Code is the power to control and regulate use; such a power is distinct from the concept of “ownership” as used in the Civil Code and in common usage. 16

The same concept has been expressed by the legislative bodies of several states. 17 For example, Mississippi Code Annotated section 51-3-1 provides in part that “. . . the general welfare of the people of the State of Mississippi *1031 requires that the water resources of the state be put to beneficial use .... All water ... is hereby declared to be among the basic resources of this state to therefore belong to the people of this state and is subject to regulation in accordance with the provisions of this chapter. The control and development and use of water . . . shall be in the state, which, in the exercise of its police powers, shall take such measures to effectively and efficiently manage, protect and utilize the water resources of Mississippi.” Although this statute more explicitly interweaves the facially inconsistent concepts of ownership and mere management than does Water Code section 102, we think there can be little doubt that the California statute has precisely the same intent and meaning as that of Mississippi.

Confirming this view is the statement in City of Long Beach v. Mansell (1970) 3 Cal.3d 462, 482 [91 Cal.Rptr. 23, 476 P.2d 423], to the effect that the State’s “ownership” of lands such as tidelands which it holds in public trust “is not of a proprietary nature.” (Italics added.) If the nature of the State’s “ownership” of waters does not even reach the level of ownership “in trust,” as is suggested by Ivanhoe Irr. Dist. v. All Parties, supra, 53 Cal.2d 692 surely it has no proprietary element at all. 18

As a point of both legal and a philosophical interest, it is debatable whether water, in its natural state, is capable of being owned and thus whether it can ever fairly be described as “property” at all. 19 (See Civ. Code, § 654.) The concept of ownership is clearly anomalous viewed against the concept of riparian rights in a flowing stream. What can a riparian landowner “own” of the waters in a flowing stream, and when and for how long can he or she own them? Even the waters of an apparently tranquil and stable lake may in fact flow in and out, or percolate from the lake boundaries and wend *1032 their way through the soil below other tracts of property. The same is true, even a fortiori, of groundwater percolating through soils. 20

Not surprisingly, therefore, it has long been held by the courts of this state that “. . . running water, so long as it continues to flow in its natural course, is not, and cannot be made the subject of private ownership.” (Kidd v. Laird (1860) 15 Cal. 161, 179-180.) Indeed, groundwater, under the absolute dominion rule, was held comparable to wild animals—ferae naturae—and was considered subject to “ownership” by the landowner only so long as it was under his land. As a bird, or deer, “belonged” to a landowner only so long as it was on his land, so was “ownership” of groundwater limited; when it flowed away, so flowed too any “ownership.” (Westmoreland N. Gas Co. v. Dewitt (1889) 130 Pa. 235 [18 A. 724].) Our Supreme Court has made the similar analogy of water to “the air, which cannot be said to be possessed or owned by any person unless it is confined within impervious walls.” (Palmer etc. v. Railroad Commission (1914) 167 Cal. 163, 168 [138 P. 997].)

The same instinctively appealing logic applies to “ownership” by the State when the essentially evanescent and/or transitory character of water in its natural state is considered. If the State can own the water of the state, what becomes of the State’s “ownership” of water in a river which crosses state or national boundaries?

For this reason alone it is impossible to accept real parties in interest’s proposal 21 that the State has an ownership interest in the “corpus” of State waters even though individual users have usufructuary rights. The ownership proposed by real parties in interest is impossible to define and virtually unrelated to the common sense of the term. Furthermore, it is contrary to explicit statements in Kidd v. Laird, supra, 15 Cal. 161, to the effect that the State does not have a property right in the “corpus” of the waters. (15 Cal. at p. 180.) We decline to recognize such a partial proprietary “ownership.”

Further Issues

In addition to its frontline attack on the trial court’s ruling that the “owned property” exclusion applies, the State also argues that even if the exclusion would otherwise apply to the groundwater, there is nevertheless *1033 coverage because the true damage is to the public interest and the contamination constitutes a threat to public health. We are not persuaded by their arguments or the authorities presented in support of them.

The State relies on Intel Corp. v. Hartford Acc. and Indem. Co. (N.D.Cal. 1988) 692 F.Supp. 1171, 1184 (affd. in part, revd. in part on other grounds, Intel Corp. v. Harford Acc. & Indem. Co. (9th Cir. 1991) 952 F.2d 1551) for the proposition that “costs incurred ... to abate this public danger must fall outside the ambit of [the “owned property” exclusion].” However, despite certain language relating to the importance of protecting public health from injury due to toxic contamination, the court essentially relied on its understanding that, under California law, the insured, Intel, did not own the contaminated groundwater under its real property. Hence, the “owned property” exclusion did not apply. The case does not support the broad proposition that the State wishes it did. Although some other cases from lower courts in other jurisdictions may contain language more directly helpful, we decline to hold, on the limited analysis provided to us, that an “owned property” exclusion somehow automatically becomes inapplicable if public health is endangered by the injury to the insured’s property. 22

As a fallback position to its challenge on the issue of “ownership,” the State also argues that the exclusion certáinly is not applicable insofar as the contamination caused injury to third party property. In its order, the trial court expressly deferred the question to phase III of the litigation. Accordingly, we need not and do not discuss it.

The last issue which we must address concerns the trial court’s finding that the “owned property” exclusion was not ambiguous. Given the tenor of our discussion in this case, it will be obvious that we cannot agree, although the nature of our analysis means that we approach the issue of ambiguity somewhat differently than does the State. 23

The State “owns” the groundwater in a regulatory, supervisory sense, but it does not own it in a possessory, proprietary sense. This being so, it is *1034 apparent that the “owned property” exclusion, which we believe we can safely say normally applies to property owned by the insured in the manner defined by the Civil Code, is not clearly applicable here. An ambiguity arises when the language of the policies is read against the complex matrix of law into which we have delved in this opinion.

We are not prepared to determine how the language in the policies should be interpreted. The principles of insurance policy construction are set out in numerous cases, including AIU Ins. Co. v. Superior Court, supra, 51 Cal.3d at pages 821-823. As these principles were never applied below, it is appropriate for us to defer to the trial court in the first instance. Furthermore, the record is not adequate for us to determine 1) who drafted the policies, 2) whether the term “owned property” was used in an “ordinary and popular” sense, and 3) what either party understood the language to mean a) in general, or b) with respect to unconfined and unappropriated waters. We therefore do not reach the issue but leave it for the trial court to resolve after such further proceedings as it and the parties may deem appropriate.

Our holding is simply that the State does not “own” the groundwater so as to make the exclusion in the policies applicable as a matter of law. We express no opinion on the correct construction of the exclusion in the context of this case, as that question depends on facts not before us.

The petition for writ of mandate is granted in part. Let a peremptory writ of mandate issue, directing respondent court to vacate its finding that the “owned property” exclusion applies in this case and to conduct further proceedings to resolve the ambiguity inherent in the policy language. Petitioner shall recover its costs.

Hollenhorst, Acting P. J., and Ward, J., concurred.

The petition of real parties in interest for review by the Supreme Court was denied June 2, 2000. Brown, J., was of the opinion that the petition should be granted.

Notes

The record submitted with the petition does not include copies of the relevant policies. A copy of a page apparently intended to be “representative” is attached to a stipulation included as exhibit 1. The language in that policy provides that it “does not apply ... to injury, [illegible] or destruction of the property owned by the State.”

No review is sought of the court’s other rulings contained in the decision.

Real parties in interest, in their joint opposition to the petition, assert that the State has failed to show the necessity for extraordinary review in this case. Our issuance of the order to show cause operates as a conclusive finding that any remedy by way of eventual appeal is not adequate here. (Robbins v. Superior Court (1985) 38 Cal.3d 199, 205 [211 Cal.Rptr. 398, 695 P.2d 695].) Accordingly, we do not revisit the issue.

The English common law rule, which still has a few adherents in this country, recognized the “absolute dominion” of a landowner over groundwater to which he had access. (3 Waters and Water Rights, supra, § 21.01 et seq., p. 115 et seq.) Notably, this dominion rested in the landowner, not the government. Under the strictest version of the rule, a landowner could pump out as much groundwater as he chose, even if by so doing he caused his neighbor’s well to dry up, or the neighbor’s overlying land to collapse. (Id. at §§ 21.02, pp. 116-117, citing and quoting from Langbrooke Props., Ltd. v. Surrey Cty. Council (1969) 3 All E.R. 1424; and Stephens v. Anglian Water Auth. (1987) 3 All E.R. 379.

At least as against competing demands.

This point is made in AIU Ins. Co. v. Superior Court (1990) 51 Cal.3d 807, 817, footnote 5 [274 Cal.Rptr. 820, 799 P.2d 1253]. We will directly address this case and its relevant language in the fourth part of our discussion.

Govemment Code section 240 defines “The People, as a political body” as “Citizens who are electors” and “Citizens not electors.”

One may also compare the nature of the State’s ownership of tidelands and navigable waters under Civil Code section 670 (quoted in part post), which is expressly held in modem cases to be in the nature of a public trust. (State of Cal. ex rel. State Lands Com. v. Superior Court (1995) 11 Cal.4th 50, 63 [44 Cal.Rptr.2d 399, 900 P.2d 648].) By contrast, the Supreme Court has repudiated the theory that the State holds water in trust for the public. (See Ivanhoe Irr. Dist. v. All Parties (1960) 53 Cal.2d 692, 716 [3 Cal.Rptr. 317, 350 P.2d 69].) We will also return to this point later.

Then and now, Water Code section 1252 authorizes “Any person . . . [to] apply for and secure from the board, in conformity with this part and in conformity with reasonable rules and regulations adopted from time to time by it, a permit for any unappropriated water.”

An example, of course, is the language from City of San Bernardino v. City of Riverside just quoted.

We repeat that once water has been appropriated in a discrete form or quantity, it can become personal property. (Lewis v. Scazighini, supra, 130.Cal.App. 722.) This is consistent with Civil Code section 655, but we interpret the statute, with respect to water, as applying to water which has been appropriated.

The governmental entities, in turn, were demanding that the policyholder, as the truly responsible party, reimburse them for these expenses. Thus, the specific question was whether the governmental expenses were “damages” suffered by a third party and therefore covered by the policies.

In Fireman’s Fund, supra, the court comments that “we are not bound by dicta’.’ and further that such statements by a higher court have “no force as precedent.” (65 Cal.App.4th at p. 1301.) Other courts have more cautiously remarked that dicta from the Supreme Court is to be carefully considered. This court has previously confirmed its independence from the control of mere dicta, but at the same time noted that dicta which reflects a “thorough analysis” or “compelling logic” should be followed. (County of San Bernardino v. Superior Court (1994) 30 Cal.App.4th 378, 388 [35 Cal.Rptr.2d 760].)

The Supreme Court found that cleanup costs for which the policyholder plaintiff had to reimburse the government were “damages” because 1) the governmental entities had suffered injury to property (the portion of the discussion relied upon by real parties in interest here) and 2) the investigation and removal expenses were “loss” due to the policyholder’s actions. The two reasons appear to be of equal dignity, although the discussion then focuses on the second prong, as the court holds th'.t such “loss" falls under the policy even if the agency incurs expense for cleaning up property in which it does not have a proprietary interest. Thus, arguably the comments about the State’s ownership interest in the water was not necessary to the decision.

Because we consider the effect of the Supreme Court’s statement to be less than it seems, we need not discuss the cases which apply it with a similar lack of analysis. (E.g., A-H Plating, Inc. v. American National Fire Ins. Co. (1997) 57 Cal.App.4th 427, 442 [67 Cal.Rptr.2d 113].) We similarly omit detailed discussion of cases which, in fact situations similar to that of AIU Ins. Co., reach similar results without delving into the historical *1030 intricacies of water law. We do note, however, that a pre-AIU Ins. Co. case which is cited by real parties does comment that the State has a “property interest in groundwater,” relying on Water Code section 102. At the same time, however, it concedes that “in this state, all ownership of water is usufructuary.” (Aerojet-General Corp. v. Superior Court (1989) 211 Cal.App.3d 216, 229 [257 Cal.Rptr. 621].) The case is therefore of little assistance to real parties, as it does not address the problem of what kind of “property” right one has in a substance as to which one’s right is also merely “usufructuary”; and see also City of San Bernardino v. City of Riverside, supra, 186 Cal. 7, in which the court noted that the State’s property rights in water arose not from section 102, but from its status as a riparian or overlying landowner.

A still less definitive analysis appears in an indemnity and contribution case between two polluters, one of which was being compelled to clean up the toxic mess. (Selma Pressure Treating Co. v. Osmose Wood Preserving Co. (1990) 221 Cal.App.3d 1601, 1615-1618 [271 Cal.Rptr. 596].) The court first set out the public trust doctrine from which the Supreme Court retreated in Ivanhoe Irr. Dist. v. All Parties, supra, 53 Cal.2d 692, and then proffered both a proprietary theory and the parens patriae approach to explain the State’s power to sue over injury to groundwater.

It is worth noting at this point that in all of the cases which purported to explain the nature of the State’s interest in groundwater, the State was not a party. Hence, there is no question of estoppel. The State did not benefit from the holdings in those cases and is not taking an inconsistent position here.

The Code Commissioner’s note to section 102 comments that pursuant to the Constitutional provision above quoted, “the State exercises governmental, rather than strictly proprietary, control over the water resources of the State.” (Code commrs. note, reprinted at 68 West’s Ann. Wat. Code (1971 ed.) foll. § 102, p. 142.)

It may be doubted whether any state does not assert the power to regulate the use of water to some extent at least. See generally 3 Waters and Water Rights, supra, especially its discussion of the erosion of the absolute dominion rule with respect to groundwater; also the discussion of legislative control over correlative and appropriative rights, especially in the *1031 imposition of requirements based on reasonable use or the avoidance of harm to other water or land interests. (Chs. 20-24, passim.)

In some cases, the State’s undoubted public trust interest in tidelands and navigable waters must be considered in connection with the statutory appropriative system of water rights. However, we do not read National Audubon Society v. Superior Court, supra, 33 Cal.3d 419 [189 Cal.Rptr. 346, 658 P.2d 709], as recognizing a public trust interest in the waters themselves. Instead, the court discusses the public trust interest of the state in tidelands and navigable waters, and explains how the interests served by the public trust must be considered—although not necessarily given precedence—when water usage is considered. (33 Cal.3d at p. 446.) For example, the state’s interest in agriculture may require that water rights be awarded with respect to a navigable waterway, even though the result will be deleterious to such public trust uses in the waterway as fishing, commerce, or even recreation; however, as the court explains, the decision should be made after considering all factors.

Philosophically the answer would seem to be “no,” but we realize that courts and legislatures can make rules independent of philosophy.

One reason for the acceptance of the absolute dominion rule (see fn. 4, ante) with respect to groundwaters was the lack of knowledge about the collection and movement of such water. Groundwater was often called “occult” (see Frazier v. Brown (1861) 12 Ohio St. 294, 311, quoted in 3 Waters and Water Rights, supra, § 20.02, p. 53) and the term, carrying the meanings of both “hidden” and “mysterious,” must have seemed particularly appropriate.

Their trial court presentation did not focus on the purported distinction we discuss here.

See Unigard Mut. Ins. Co. v. McCarty’s, Inc. (D.C. Idaho 1988) 756 F.Supp. 1366, 1368-1369, for a somewhat convoluted analysis in this respect. We do not mean to suggest, however, that we disagree with the result. Insofar as there is injury to the property of another, coverage may exist. As we note below, this issue is not properly before us.

The State’s argument in its briefs begins from the assumption that we will find that it is “the owner” of the groundwater and that the exclusion therefore applies. Its position is that the “owned property” exclusion contains a latent ambiguity because, in the circumstances of this case, injury to the State’s property (if the State is held to be the owner) also damages other property interests. As we have noted above, the trial court generally deferred consideration of the extent to which third party interests are in fact involved and as to which coverage may therefore apply.