concurring.

The state appeals a pretrial order suppressing evidence in this criminal proceeding. ORS 138.060. Defendant was charged with possession of a controlled substance, ORS 475.992(4)(a), and moved to suppress evidence found in his home by Lake Oswego police officers who had responded to a burglar alarm. The trial court granted the motion on the ground that the officers’ entry into defendant’s home violated defendant’s rights under Article I, section 9, of the Oregon Constitution and the Fourth Amendment to the United States Constitution. For the reasons stated below, I would conclude that the proper disposition of the appeal in this case is to affirm the trial court.

This court reviews suppression decisions for legal error and is bound by the trial court’s findings of historical fact if they are supported by evidence in the record. State v. Ehly, 317 Or 66, 74-75, 854 P2d 421 (1993). To the extent that the trial court did not make express findings, we resolve disputed facts in a way that is consistent with its ultimate conclusion. Ball v. Gladden, 250 Or 485, 487, 443 P2d 621 (1968).

The following facts are taken from the trial court’s findings, as well as from the record made at the suppression hearing. On February 23, 2002, at approximately 6:30 p.m., the burglar alarm at defendant’s house went off and two of defendant’s neighbors reported it to the police. One of the neighbors told the dispatcher that defendant was the only one who lived at the house and that defendant had left several hours earlier to play in a basketball game in Portland. Officer Brady responded about 15 minutes later. It was dark when he arrived. Brady spoke with one of the neighbors who had called about the alarm and that neighbor reported that he had seen a car that he did not recognize back out of defendant’s driveway 45 minutes to an hour earlier. The neighbor also reported that he had kept an eye on defendant’s house after calling the police. Brady observed that the door to defendant’s house was open about a foot. Brady walked around the exterior of the residence looking for signs of forced entry and attempting to determine if anyone was inside. He *401saw no sign of forced entry or of anyone inside. Officer Forman arrived approximately 15 minutes after Brady.

The officers then entered and searched defendant’s house pursuant to a Lake Oswego Police Department policy on responding to burglar alarms. That policy required them to check anywhere that a person could reasonably hide. Pursuant to that policy, the officers looked in closets and cabinets, behind couches, and under desks. While searching the upstairs area of the house, Brady saw a small door that led to an attic storage area. Brady opened the door and saw a large plastic bag that appeared to contain several smaller bags of marijuana. He also noticed an odor of marijuana as he opened the door. The officers confiscated the large bag of marijuana and then completed their search of defendant’s house. In a bedroom, they discovered two partially smoked marijuana cigarettes, which they did not confiscate. They found no evidence of burglary. The officers left a false-alarm notice for defendant, indicating that they had been in the house and informing him why they had been there.

Brady returned to defendant’s house approximately a week later and told defendant about his discovery of the bag of marijuana. Defendant stated that the marijuana belonged to a friend. He further stated that he had removed a small amount of marijuana from one of the bags for personal use.

Defendant was charged with possession of a controlled substance. ORS 475.992(4)(a). He moved to suppress the evidence found in his home on the ground that it was obtained in the course of an illegal search. Defendant argued that the search was warrantless and did not fall within any exception to the warrant requirements under either Article I, section 9, of the Oregon Constitution or the Fourth Amendment to the United States Constitution. In response, the state argued that three exceptions to the warrant requirement justified the search. First, the state argued that defendant consented to the search by implication because he had a burglar alarm system and had contracted with an alarm company to attempt to notify the police when the alarm was activated. Second, the state posited that the search was a valid administrative search. Finally, the state asserted that the officers had probable cause to believe that a crime was being *402committed and that exigent circumstances obviated the need to obtain a warrant.

At the suppression hearing, the parties agreed that, between 1996 and 2002, there had been 32 false alarms at defendant’s house. Further, defendant had been cited approximately seven times under a Lake Oswego false alarm ordinance. The trial court found, and the state does not dispute on appeal, that over 99 percent of all residential alarms that came to the attention of the Lake Oswego police during the year in which defendant’s house was searched were false alarms. Officer Brady’s testimony indicated that false alarms often occurred when a homeowner set off an alarm while leaving the house. The trial court concluded that, given the high rate of false alarms in Lake Oswego, “a sounding alarm cannot add to the showing of probable cause.”

The state also offered evidence showing that Lake Oswego has an ordinance that imposes civil penalties after a certain number of false alarms and argued that the challenged search constituted a valid administrative investigation into a possible false alarm. Finally, the state offered evidence about the manner in which such alarms were investigated, specifically describing the policy of the Lake Oswego Police Department concerning the scope and intensity of searches when an alarm has been set off and the premises are not secure. The evidence also indicated that this policy was consistently followed by the department.

The trial court ultimately rejected each of the state’s arguments as to why the search was valid despite the lack of a warrant. First, the trial court concluded that the evidence did not establish in a clear and unambiguous manner that defendant impliedly consented to a search of his house of the scope and intensity of the search that the police conducted. Second, the trial court rejected the state’s administrative search argument on the ground that an extensive search of a private residence, conducted in order to investigate possible violations of a civil ordinance, went far beyond the types of administrative searches that courts have traditionally recognized as valid. Finally, the trial court determined that, although Officer Brady subjectively believed that he had probable cause to enter defendant’s home, that belief was not *403objectively reasonable under the circumstances. The trial court identified three circumstances relevant to objective probable cause: the sounding of the alarm, the partially open front door, and the unidentified car leaving defendant’s property approximately an hour before the police arrived. The court concluded that the sounding alarm did not contribute to objective probable cause because the false alarm rate in Lake Oswego is so high. The court also noted that no testimony established the existence of any exigent circumstances that required the officers to enter immediately rather than seek a warrant and concluded that the possibility that an intruder was present in the house did not, by itself, create an exigency. Accordingly, the trial court granted defendant’s motion to suppress the evidence obtained from the search of his house.

The state appeals the trial court’s suppression of the evidence. ORS 138.060(l)(c). It reiterates the arguments that it made in the trial court: that probable cause and exigent circumstances justified the search; that defendant consented to the search; and that the search was a valid administrative search.

Before turning to the issues raised by the state on appeal, we must address the significance of the state’s failure to challenge the trial court’s conclusion that the search of defendant’s home violated the Fourth Amendment. In the trial court, defendant made extensive and distinct arguments in support of suppression under both Article I, section 9, and the Fourth Amendment. The trial court explicitly concluded that the search violated both Article I, section 9, and the Fourth Amendment. On appeal, the state’s brief cites neither the Fourth Amendment nor any of the relevant case law interpreting it. In its discussion of probable cause and exigent circumstances, the state’s brief specifically addresses only Article I, section 9. Nothing in the state’s brief even suggests that the state was aware of the court’s holding under the Fourth Amendment. It is axiomatic that, when a trial court bases a decision on multiple grounds, an appellant may prevail on appeal only after demonstrating that all of the bases for the court’s decision were erroneous. See Roop v. Parker Northwest Paving Co., 194 Or App 219, 236, 94 P3d 885 (2004) (“where [appellants] fail to challenge the alternative basis of the trial court’s ruling, we must affirm it”).

*404Several months after the argument in this case, this court asked the parties to submit supplemental memoranda discussing whether the state’s appeal adequately challenged both of the bases for the trial court’s suppression ruling. Defendant argues that, because the state did not make any arguments about the Fourth Amendment in its opening brief, the correct resolution of this case is for this court to affirm the trial court based on its unchallenged holding that the search violated the Fourth Amendment. The state responds that it is clear from the context of the parties’ arguments in the trial court that both parties believed that the Oregon Constitution provided at least as much protection as the federal constitution, and that a search that satisfied state constitutional standards necessarily satisfied federal constitutional standards. Thus, the state argues, “the Oregon constitutional issues raised by defendant are dispositive of his federal constitutional claims and thus no separate federal analysis is necessary to address the issue presented here.”1

I would conclude that the state’s argument is fundamentally flawed. The first flaw with it is that it came too late. *405Before it responded to our post-argument questions, the state had made no argument to us about the Fourth Amendment.2 Even if the state’s argument in response to our questions were correct, the argument had to be made before the case was submitted to us for decision. Because it was not, we cannot now consider and adopt the argument as the basis for our conclusion that the trial court erred in suppressing the evidence under the Fourth Amendment.

The second flaw with the state’s argument is that it ignores more than 20 years of Oregon case law that has established that the state and federal constitutions are to be interpreted and enforced independently. Independent analysis may lead to the same result in many cases, but this court is required to engage in the analysis. Rather than engage in the analysis, the state relies on the syllogism that Oregon constitutional law on probable cause and exigent circumstances is at least as protective as federal constitutional law, the search in this case satisfied Oregon law, therefore it satisfied federal law. This court cannot use that syllogism to avoid its obligation to make an independent assessment of probable cause and exigent circumstances under both bodies of law.3

*406The state places significant reliance on State v. Greene, 285 Or 337, 591 P2d 1362 (1979), as support for the proposition that Article I, section 9, is at least as protective as the Fourth Amendment, which is an essential step in the syllogism on which it relies. That reliance is problematic. Greene was based on the principle that Article I, section 9, is to be interpreted consistently with the Fourth Amendment, absent a reason to depart from the Fourth Amendment analysis. 285 Or at 339. Under that approach, it would necessarily follow that Article I, section 9, would be at least as protective as the Fourth Amendment.

Critically, however, the court repudiated that approach three years later in State v. Caraher, 293 Or 741, 653 P2d 942 (1982). Caraher is the case that established that Article I, section 9, is to be interpreted independently of the Fourth Amendment and that began 20 years of independent development of Oregon search and seizure law. Independent development of the law under Article I, section 9, can lead to situations in which that law is less protective than is the law under the Fourth Amendment. See, e.g., Bank of Oregon v. Independent News, Inc., 298 Or 434, 693 P2d 35 (1985) (Article I, section 8, less protective than First Amendment on aspect of defamation law); State v. Flores, 68 Or App 617, 625, 685 P2d 999, rev den, 298 Or 151 (1984) (independent analysis of Article I, section 9, can lead to less protection against search and seizure than provided by Fourth Amendment). Consequently, Greene can no longer be considered to be authority for the principle that Article I, section 9, is as protective of individual rights as is the Fourth Amendment.

Because of the change wrought by Caraher and similar cases, this court’s task is to determine independently whether a particular search violated Article I, section 9, and whether it violated the Fourth Amendment. It cannot short circuit that task by doing what Judge Deits’s dissent does *407here. It looks to see if there is existing case law that establishes that Article I, section 9, involves a different analysis or produces a different result on a particular legal issue than does the Fourth Amendment. 198 Or App at 423-24 (Deits, J. pro tempore, dissenting). If, as here, no one has identified a case or advanced an argument that establishes a difference between the two provisions, then they are the same and produce the same result, and a decision under one of them will determine the decision under the other. Hence, according to Judge Deits, a conclusion that the court erred in suppressing the evidence under Article I, section 9, is sufficient to establish that the court erred in suppressing the evidence under the Fourth Amendment. Put simply, that is not how Oregon’s independent analysis works.

Judge Deits’s approach flies directly in the face of the logic on which Oregon’s independent analysis is based. This case can illustrate the point. Assume that this court accepted Judge Deits’s position that the assessment of objective probable cause and exigent circumstances is the same under Article I, section 9, and the Fourth Amendment and that it reversed the trial court’s suppression order on that basis. Assume that the next case that the court decides involves only an application of Article I, section 9, and it concludes that a particular search did not involve exigent circumstances. Assume further that, on the same day and on identical facts, the United States Supreme Court concludes under the Fourth Amendment that the equivalent search did involve exigent circumstances. Under the logic of independent analysis, both decisions would be correct, but that would mean that our decision in this case would no longer be correct. The point of independent analysis is that the two provisions are not linked, so there is no need to determine, as Judge Deits dissent does here, if they can be. By linking them as Judge Deits’s does, she would create a precedent that would almost certainly have to be overruled in the future, not because it is necessarily wrong to conclude that the state and federal analyses lead to the same result at this point in the independent development of state and federal search and seizure law, but because the inevitable development of case law *408by the state and federal courts will cause the state and federal law to drift apart.4 Proper application of Oregon’s independent analysis avoids the problem that Judge Deits would create in this case, a problem that she would create in an effort to find a way out of the box that the state created for itself by focusing exclusively on Article I, section 9, in its challenge to the trial court’s suppression order.5

Even if the state’s failure to challenge the trial court’s Fourth Amendment ruling did not foreclose this court’s ability to overturn the trial court’s suppression order, I would nevertheless affirm the suppression order on its merits. I would do so because I am persuaded that the trial court correctly rejected the exceptions that the state advanced to support the warrantless search of defendant’s home.

I first consider the exception based on probable cause and exigent circumstances. In assessing probable cause, this court considers the totality of the circumstances. See, e.g., State v. Kappel, 190 Or App 400, 404, 79 P3d 368 (2003), rev den, 336 Or 509 (2004). Viewed in that light, the *409evidence shows that there was only a remote possibility that the officers would find a burglar or evidence of a burglary in the house, and that falls well short of probable cause.

The trial court found that there had been 32 false alarms at defendant’s home over a seven-year period. Furthermore, during the year in which the police searched defendant’s home, over 99 percent of all residential alarms in the city were false alarms. Those facts provide useful background in assessing the significance of the events that occurred the night that the police searched the home.6

The officers arrived at defendant’s home in response to telephone calls from two neighbors reporting an alarm at the house. One of the neighbors told them that defendant lived alone in the house and had left several hours earlier to play in a professional basketball game in Portland. The neighbor said that he had watched the house after the alarm sounded and had not seen anyone moving around it. Finally, the neighbor said that 45 minutes to an hour before the alarm sounded he had seen an unfamiliar car back out of the driveway. The officers walked around the house looking for evidence of forced entry but found none. They did find the *410front door ajar about one foot, but they did not testify about whether the door lock had been set. There were no other doors or windows open and no signs of forced entry.

Given the evidence, it is apparent that the alarm could have been triggered only by the opening of the front door. Because there was no sign of forced entry and there was no other unsecured door or window, it also is evident that, if a burglar opened the front door, the burglar did it from the outside rather than the inside. Further, because the car that drove away from the house left more than 45 minutes before the alarm sounded, the person or people in the car presumably did not cause the alarm to sound by opening the door.7

The issue, then, is whether it is probable that the front door was open because a burglar opened it. Given the absence of evidence of a forced entry, the number of false alarms that defendant had experienced, and the overall ratio of false alarms in the city, I believe that it is much more likely that the front door was accidently left ajar by defendant and opened on its own.8

Even assuming that it is probable that a burglar entered the open front door, it is improbable that the police would have found a burglar or evidence of a burglary in a search of the house. That is because the alarm would have scared off the person when it sounded. Significantly, one of the neighbors told the officers that he had watched the house *411after the alarm had gone off and had not seen anything or anyone that caused him to be concerned.

Judge Deits implicitly assumes that it is probable that a burglar chose to enter the house notwithstanding the sounding alarm and notwithstanding the fact that there were neighbors living nearby who could see the house and people, if any, moving around it. She necessarily also concludes that it is probable that the burglar decided to hide in the house while the alarm sounded, because the neighbor said that he had not seen any activity around the house after the alarm went off and the police did not see anything unusual in the house when they looked through the windows.

It certainly is possible that a burglar triggered the alarm by opening the front door and leaving it open, that the burglar entered the house notwithstanding the sounding alarm and the presence of neighbors, that the burglar stayed hidden in the house for an extended period of time, and that the officers could have found the burglar or evidence of the burglar’s activities when they entered the house. That possibility is quite remote, however. Consequently, it is improbable that the police would find a burglar or evidence of a burglary when they entered defendant’s house. Given the facts of this case, no court would have authorized a warrant to search the house, because no court could have concluded that there was probable cause to believe that a burglary was in progress or had occurred.

Even assuming that there was probable cause to believe that the police would find a burglar or evidence of a burglary in defendant’s house, there were no exigent circumstances that would justify the decision by the police to enter defendant’s home to conduct a search without first obtaining defendant’s consent or a warrant. The police knew that defendant was in Portland playing in a professional basketball game. He could have been reached by telephone to ask for his consent to enter his home. Contacting defendant by telephone would also have given the police an opportunity to learn more about what had caused the alarm to sound.

Judge Deits nevertheless concludes that an exigency was created by the possibility that the putative burglar would escape before the police could get a warrant. She says *412that an immediate entry was necessary to prevent escape by the burglar or damage to defendant’s property. 198 Or App at 430-32 (Deits, J. pro tempore, dissenting). She rejects the idea that the officers could have called for backup officers to surround the house while they sought a warrant. But contrary to Judge Deits’s reasoning, the likelihood of escape would have been reduced if the house had been surrounded by officers when the search occurred. Before entering the house, the officers had walked around it looking for signs of forced entry, had checked doors and windows for signs that someone had broken in, and had shone flashlights into the house. If a burglar were in the house, the burglar would likely have become aware of that activity and been prepared to respond if the officers entered the house. When the officers entered the front door of the house and announced their presence, the burglar could well have decided to leave the house through an unguarded door or window, thereby avoiding capture. As for the risk of damage to defendant’s property, the officers had seen none and there is no reason to believe that a burglar would begin damaging property while police surrounded the house and waited for a search warrant.

The more fundamental problem with Judge Deits’s position is that it mistakenly assumes that securing a warrant could have required at least one of the officers to leave the house for a significant time. Id. at 430 (Deits, J. pro tempore, dissenting). Warrants are available by telephone, ORS 133.545(5) and ORS 133.555(3), and the availability of such a warrant is to be considered in determining the existence of exigent circumstances. See Stevens, 311 Or at 129-30 (citing State v. Wise, 305 Or 78, 82 n 3, 749 P2d 1179 (1988)). Here, the state had the burden of establishing the existence of exigent circumstances, and it offered no evidence on the availability of a telephonic warrant.

Consequently, Judge Deits is simply wrong to say that the officers were faced with two unacceptable choices if they sought to get a warrant to authorize their search of defendant’s home for the putative burglar: They would both have to leave the home in order to get the warrant, thereby permitting the burglar to escape, to damage defendant’s property, or to confront defendant if defendant returned home while they were getting the warrant. Alternatively, one *413of them would have to leave the home to get the warrant, thereby leaving the other officer without a backup and making it easier for the burglar to escape, to damage defendant’s property, or to harm the remaining officer. 198 Or App at 430-32 (Deits, J. pro tempore, dissenting). Because a telephonic warrant could be obtained without either officer leaving defendant’s home, none of the circumstances that Judge Deits identifies created an exigency that allowed the officers to search the home for a burglar without getting a warrant.

The claimed exigency also cannot be reconciled with the reason given for entering the house. The officers did not consider whether it would be feasible to obtain a warrant or whether they faced circumstances that required them to act without one. They entered the house because a Lake Oswego Police Department policy directed them to enter a house under the circumstances of this case to conduct an administrative search to determine if the alarm was a false alarm. The policy requires a police officer who responds to a residential alarm and finds an open or unlocked window or door to call for a backup officer and then enter the residence with the other officer to look for an intruder or evidence of an intrusion. If they do not find evidence that someone entered the residence unlawfully, they are to leave a false alarm notification in the residence and secure it. That is what the officers did here. There may be a categorical exigency for the search of automobiles under certain circumstances, see, e.g., State v. Kock, 302 Or 29, 33, 735 P2d 1285 (1986) (“Searches of automobiles that have just been lawfully stopped by police may be searched without a warrant and without a demonstration of exigent circumstances when police have probable cause to believe that the automobile contains contraband or crime evidence.”), but there is no categorical exigency for the search of houses. This case would create one, however, if, as Judge Deits concludes, the reasons giving rise to an administrative search were exigent circumstances justifying a warrantless search of defendant’s home.

In summary, the trial court did not err in concluding that the warrantless search of defendant’s home was not a lawful search under Article I, section 9, as a search based on probable cause and exigent circumstances. Because the state advances two alternative rationales for the search, I turn to *414them to determine if either provides a basis to uphold the legality of the search.

The state contends that the search was lawful as an administrative search to enforce a Lake Oswego ordinance that imposes a civil fine on people who have a sequence of residential false alarms. As described above, the administrative search policy directs police officers to enter and search homes in which alarms have sounded if they find an open or unlocked window or door. The officers are directed to look in every place in the home in which an intruder might reasonably be expected to hide. If the search discloses no evidence of an intrusion, then the officers are to leave a notice of a false alarm in the home, secure the home, and leave.

There are three requirements for a lawful administrative search under Article I, section 9. First, the search must be authorized by a politically accountable lawmaking body. Weber v. Oakridge School District 76, 184 Or App 415, 435, 56 P3d 504 (2002), rev den, 335 Or 422 (2003). Second, it must be designed and systematically administered so that it involves no discretion by the officers conducting the search. Id. at 436. Third, it must be reasonable in relation to its purpose. Id. at 437. The administrative search in this case runs afoul of the first and third requirements and, hence, violates Article I, section 9.

A politically accountable lawmaking body adopted the city ordinance that authorizes the imposition of a civil fine for a third false residential alarm in a one-year period and that authorizes the police department to “[investigate and verify whether an alarm is a false alarm.” Significantly, however, the ordinance does not expressly authorize the police to conduct searches of residences in order to investigate and verify whether alarms are false. Although the authority to conduct an administrative search can be implied from a grant of related authority, it cannot be implied here. The instances in which authority has been implied have involved situations in which the administrative searches were necessary to accomplish the tasks that the granting authority expressly authorized. See, e.g., State v. Boone, 327 *415Or 307, 314, 959 P2d 76 (1998) (authority to inventory vehicles implied from explicit authorization by city ordinance to impound them); State v. Ketelson, 163 Or App 70, 986 P2d 1202 (1999) (authority to conduct inventory at detoxification center implied from decision by politically accountable body to establish the center). That is not the case here. It is not necessary to search homes for evidence of false alarms in order to enforce an ordinance that imposes fines for successive false alarms. Consequently, a decision to adopt an ordinance that imposes a fine for successive false residential alarms does not imply authority for the police to conduct administrative searches of residences to determine if alarms are false.

Even if the authority could be implied, it would not be reasonable to search people’s homes as a means of enforcing an ordinance that imposes a civil fine for successive false alarms. A search of a private home for evidence of intrusion is in itself a dramatic intrusion into the privacy of the resident. That intrusion cannot be justified when viewed in light of the goal of the ordinance and the lack of a need for the intrusion to achieve it.

The state also contends that defendant impliedly consented to the search by installing an audible burglar alarm in his house. He did not. A person who installs such an alarm presumably understands that neighbors might call the police when the alarm sounds and that the police might enter the house under appropriate circumstances in response to the alarm, but the person does not consent in advance to every entry by the police in response to an alarm. Furthermore, the police did not enter the house because they believed that they had defendant’s consent to do so. They entered because they were directed by a police department policy to determine if the alarm was a false alarm. The search cannot be upheld as a consensual search.

For the foregoing reasons, I conclude that the trial court’s suppression order should be affirmed.

Schuman and Ortega, JJ., join in this concurrence.

concurring.

The trial court’s decision plainly was based on the conclusion that the officers’ entry into defendant’s home violated his rights under Article I, section 9, of the Oregon Constitution and under the Fourth Amendment to the federal constitution. On appeal, the state challenged only the state constitutional basis for the trial court’s decision and ignored the federal constitutional holding. I am not persuaded by Judge Deits’s contention that, by citing cases that either happened to mention the federal constitution or happened to refer to other cases that addressed a federal constitutional issue, the state adequately challenged the federal constitutional basis for the trial court’s decision. Nor am I persuaded, for the reasons articulated in Judge Armstrong’s opinion, by Judge Deits’s other arguments in support of the same conclusion.

The state frequently — and successfully — complains when defendants fail to challenge all possible grounds for a trial court’s decision, arguing that the failure to address an alternate ground for decision fails to demonstrate reversible error. See, e.g., State ex rel SOSCF v. Duncan, 164 Or App 610, 612, 993 P2d 818, (1999), rev den, 330 Or 361 (2000) (“[B]ecause mother’s challenge is directed to only one of two grounds on which the termination order is based, we affirm.”). And correctly so. See, e.g., Jensen v. Medley, 336 Or 222, 239-40, 82 P3d 149 (2003) (although trial court’s jury instruction relating to one of the plaintiffs theories of liability was erroneous, court affirmed the verdict because there was another basis for it that the defendant did not challenge on appeal); Roop v. Parker Northwest Paving Co., 194 Or App 219, 236, 94 P3d 885 (2004), rev den, 338 Or 374 (2005) (“[W]here plaintiffs fail to challenge the alternative basis of the trial court’s ruling, we must affirm it.”); State v. Branstetter, 181 Or App 57, 62 n 3, 45 P3d 137, rev den, 334 Or 632 (2002) (“ ‘[TJhere is no authority for the proposition that, without invoking “plain error,” we can reverse the trial court on grounds not argued to it.’ Much less is there any authority that we can reverse on grounds not argued to us.” (Emphasis in original; citation omitted.)). The state failed to challenge all possible grounds for the trial court’s decision in this case. Sauce for the goose, sauce for the gander.

*417Judge Haselton insists that the state’s failure to address the federal constitutional justification for the trial court’s decision is of no moment, because, under State v. Kennedy, 295 Or 260, 666 P2d 1316 (1983), the trial court’s decision “was procedurally improper — or, more accurately, not a ‘holding’ at all.” 198 Or App at 441 (Haselton, J., dissenting). According to Judge Haselton, because of the “first things first” precept — requiring adjudication of state constitutional claims before reaching federal claims — the trial court’s pronouncement on the Fourth Amendment cannot be regarded as an independent ground for its decision, because the court lacked authority even to speak to the issue. Id. at 441-42 (Haselton, J., dissenting).

I do not question the appropriateness of deciding “first things first,” although, as I have noted in previous cases, the appellate courts of this state appear somewhat less than consistent in their application of the principle. See, e.g., Coast Range Conifers v. Board of Forestry, 189 Or App 531, 538-39, 76 P3d 1148 (2003), rev allowed, 337 Or 476 (2004) (“Since the early 1980s, however, the courts have been less consistently committed to the first-things-first doctrine.”). I do not agree, however, that the doctrine somehow precludes state courts from doing precisely what the court did in this case. I arrive at that conclusion for two reasons.

First, Judge Haselton’s premise is erroneous. His argument holds together only if it is assumed not merely that courts should not address federal arguments once it has been determined that state law affords a remedy but that they cannot do so. In point of fact, Kennedy says nothing about that. Indeed, Kennedy stands for the much more limited proposition that the failure of the parties to brief an issue of state law cannot preclude appellate courts from addressing the issue. Kennedy, 295 Or at 268. That is a far cry from holding that trial courts are without authority to address both federal and state law issues that, in fact, were fairly presented to those courts.

As it turns out, there is a different case, Sterling v. Cupp, 290 Or 611, 614, 614 n 2, 625 P2d 123 (1981), that does say, as a matter of federal constitutional law, that one of the justifications for a first-things-first doctrine is the notion *418that, until it is determined that state law affords no remedy, there is no violation of federal law. Since that time, however, the United States Supreme Court has held that — at least as to nonprocedural federal constitutional guarantees — as a matter of federal law, “the [federal] constitutional violation * * * is complete when the wrongful action is taken,” Zinermon v. Burch, 494 US 113, 125, 110 S Ct 975, 108 L Ed 2d 100 (1990), not when a state court decides that state law does not afford a remedy. Thus, while there may be — and, indeed, are — many reasons for adhering to a first-things-first doctrine as a prudential matter, see, e.g., Wallace P. Carson, Jr., “Last Things Last”: A Methodological Approach to Legal Argument in State Courts, 19 Willamette L Rev 641, 647-50 (1983) (history, logic, stability, and efficiency support policy of addressing state constitutional law before federal), it is perhaps no longer tenable to suggest that state courts are without authority to reach federal law issues if state law affords relief.

Second, even assuming the correctness of Judge Haselton’s premise, his conclusion does not necessarily follow from it. Merely because the courts are enjoined to address state constitutional issues first does not prevent a court, once having decided a case on state constitutional grounds, from determining that there is an independent federal constitutional basis for the decision, should its state court justification be incorrect. Such determinations are no less “holdings” than any other alternative grounds for trial court’s decisions. Indeed, they are a common practice in the trial courts and are firmly rooted in considerations of judicial economy. Judge Haselton himself appears to acknowledge that his reading of the first-things-first doctrine necessarily would force trial courts to decide cases “in piecemeal fashion, which might result in an otherwise unnecessary remand, seriatum appeals, and delay[.]” 198 Or App at 441 n 1 (Haselton, J., dissenting). It takes little imagination to realize the havoc that we would wreak upon our system by requiring trial courts to decide each issue upon one — and only one — basis at a time.

I therefore adhere to my conclusion that this case should be affirmed on the ground that the state did not challenge on appeal. In saying that, however, I express no opinion *419about the merits of either the state or the federal constitutional issues that Judge Deits and Judge Armstrong debate.

dissenting.

This case concerns the trial court’s suppression of evidence found by police in defendant’s home during the investigation of a burglar alarm. For the reasons set forth below, I would reverse and remand this case.

The pertinent facts are set forth in Judge Armstrong’s concurrence. To summarize, police responded to reports of a burglar alarm sounding at defendant’s Lake Oswego residence. When the police arrived, they found the front door of defendant’s home ajar, and discovered from the neighbors that defendant had left home several hours earlier to play in a basketball game, long before the alarm sounded. They also discovered that an unfamiliar car had been seen leaving defendant’s residence some time after defendant had left. The police found no external signs of forced entry or any external indication that anyone was inside defendant’s residence. The police conducted a limited search inside the open residence for intruders pursuant to department policy on responding to burglar alarms. In the course of that limited search, the police discovered a significant amount of marijuana.

Defendant was charged with possession of a controlled substance and moved to suppress on the ground that it was obtained in the course of an illegal search. Defendant argued that the warrantless search did not fall within any exception to the warrant requirements under either Article I, section 9, of the Oregon Constitution or the Fourth Amendment to the United States Constitution. The state responded that defendant consented to the search by implication because he installed a burglar alarm system and contracted with an alarm company to attempt to notify the police when the alarm was activated, that the search was a valid administrative search, and that the officers had probable cause to believe that a crime was being committed and that exigent circumstances obviated the need to obtain a warrant. The trial court suppressed the evidence on the ground that, *420because of the large number of false burglar alarms in Lake Oswego, “a sounding alarm cannot add to the showing of probable cause.” The court further rejected the state’s arguments that defendant consented to the search and that the search was a valid administrative search.

The state appeals, challenging the trial court’s conclusions. I would not reach the issues of whether defendant consented to the search or whether it was a valid administrative search because I agree with the state that the search was justified by probable cause and exigent circumstances.

In his supplemental brief in response to the court’s questions concerning whether the state had challenged the trial court’s Fourth Amendment ruling on appeal, defendant argues that, because the state did not make any arguments specifically concerning the Fourth Amendment in its opening brief, the correct resolution of this case is for this court to affirm the trial court’s ruling based on its unchallenged holding that the search violated the Fourth Amendment. The state responds that it is clear from the context of the parties’ arguments in the trial court that both parties believed that the Oregon Constitution provided at least as much protection as the federal constitution, and that a search that satisfied state constitutional standards necessarily satisfied federal constitutional standards. Thus, the state argues, “[T]he Oregon constitutional issues raised by defendant are dispositive of his federal constitutional claims and thus no separate federal analysis is necessary to address the issue presented here.”1

For the reasons explained below, I would hold that the state sufficiently raised both state and federal constitutional issues in its opening brief to this court. Contrary to Judge Landau’s assertion in his concurrence, this conclusion is not based solely on the fact that the state relied on cases that cited both the pertinent state and federal constitutional provisions. Rather, my conclusion is based on a number of *421considerations. First, the state’s assignment of error is broad enough to encompass all bases for the trial court’s decision. The state’s assignment of error states:

“The circuit court erred in concluding that the police lacked probable cause and exigent circumstances to search defendant’s house without a warrant; erred in concluding that defendant had not impliedly consented to police entry of his house; and erred in finding that the police entry was not a valid administrative search. Based on these erroneous conclusions, the circuit court erroneously entered an order granting defendant’s motion to suppress as follows:

“Tt is hereby ordered that defendant’s Motion to Suppress be granted. The basis for the court’s ruling is set forth in the court’s Findings of Fact and Conclusions of Law, which are incorporated into this Order by reference.’ ”

Ideally, in its discussion of the assignment of error, the state would have specifically cited both Article I, section 9, and the Fourth Amendment. Under the particular circumstances, however, I do not believe that the state’s failure to cite the Fourth Amendment should result in affirming the trial court’s decision on a basis that not even defendant has asserted in his respondent’s brief or at oral argument.

Further, although the trial court cited the Fourth Amendment in its letter opinion and in its detailed findings of fact and conclusions of law, it made no separate analysis under the two provisions. The court made the following findings of fact and conclusions of law under the heading “The Probable Cause/Exigent Circumstances Exception”:

“Findings of Fact

“12. Both officers testified and both had difficulty in identifying facts that gave rise to probable cause to enter and search defendant’s home;

“13. Officer Brady subjectively believed that there was probable cause to enter defendant’s home. However, he did not enter pursuant to that belief but rather pursuant to the City Policy;

“14. Lieutenant Forman did not subjectively believe that there was probable cause to enter defendant’s home. *422Rather, he believed that there was a possibility that evidence of a crime would be discovered;

“15. There were three circumstances that arguably contributed to a showing of objective probable cause to enter defendant’s home: a partially open front door, an unidentified car leaving defendant’s property approximately one hour before the police arrived, and an audible alarm;

“16. The audible alarm does not contribute to the showing of objective probable cause, for two reasons:

“a) In any house with an alarm system, that alarm will go off if a door is left open after the alarm has been activated. Therefore, the sounding of an alarm furnishes no additional information beyond the fact that a door is open;

“b) The false alarm rate in Lake Oswego is so high that a sounding alarm cannot add to the showing of probable cause;

“17. Lieutenant Forman testified that, assuming he was presented with the same circumstances as were present that evening, with the exception of the alarm sounding, he would simply close and secure the door and that would be the end of the matter;

“18. There was no one who testified to circumstances that required the officers to act immediately rather than to seek a search warrant. Officer Brady’s entry was not made pursuant to a subjective belief that such circumstances existed, but rather pursuant to the City Policy. The mere possibility that there might be an intruder present in the house does not create, by itself, a situation requiring immediate action;

“Conclusions of Law

“19. Officer Brady had subjective probable cause to enter and search defendant’s home. Lieutenant Forman did not have subjective probable cause to enter and to search defendant’s home;

“20. Officer Brady and Lieutenant Forman lacked objective probable cause to enter and to search defendant’s home;

*423“21. There were no exigent circumstances that justified the warrantless entry into and search of defendant’s home[.]”

Notably, there is no distinction either in the court’s findings of fact or in its conclusions of law between probable cause and exigent circumstances under Article I, section 9, and the Fourth Amendment. Further, and significantly, neither party treats the analysis under the two constitutional provisions as being any different as to objective probable cause and exigent circumstances. Both in briefing before this court and at oral argument, defendant offers no suggestion that the analyses under Article I, section 9, and the Fourth Amendment differ in any way that is significant to this case. Judge Armstrong, in his concurrence, suggests that Article I, section 9, and the Fourth Amendment implicate fundamentally different analyses. 198 Or App at 405-07 (Armstrong, J., concurring). While it is certainly true in theory that not all Article I, section 9, analyses are identical to Fourth Amendment analyses, the fact remains that, as applied to this case, neither the trial court’s analysis nor either party’s arguments suggests that the analysis implicated here is any different under Article I, section 9, and the Fourth Amendment.2

Accordingly, it is not unreasonable to treat the state’s argument that the trial court erred in its determination that the police lacked probable cause and exigent circumstances to conduct a warrantless search as encompassing both constitutional analyses. That characterization of the state’s argument is supported by the fact that many of the cases cited by the state on appeal are, in fact, firmly rooted in Fourth Amendment law. For example, the state relies on State v. Apodaca, 85 Or App 128, 735 P2d 1264 (1987), in which this court, without specific reference either to Article I, section 9, or the Fourth Amendment, discussed both probable cause and exigent circumstances. In Apodaca, we relied not only on our precedent but also on cases decided by the United States Supreme Court and various federal courts that were *424based on Fourth Amendment analyses. See id. at 131-33. The state also relies on cases such as State v. Stevens, 311 Or 119, 806 P2d 92 (1991), and State v. Bridewell, 306 Or 231, 759 P2d 1054 (1988), neither of which drew distinctions in analyzing state and federal probable cause and exigent circumstances questions, and both of which relied on State v. Miller, 300 Or 203, 709 P2d 225 (1985), cert den, 475 US 1141 (1986). Stevens, 311 Or at 129-30; Bridewell, 306 Or at 235. Miller, in turn, discussed only United States Supreme Court case law in deciding a question concerning the emergency exception to the warrant requirement. Miller, 300 Or at 299-300.

The state also cites State v. Greene, 285 Or 337, 591 P2d 1362 (1979), which, in addressing exigent circumstances, specifically relied only on Fourth Amendment precedent, noting that, although Article I, section 9, and the Fourth Amendment both offered protections from unreasonable searches and seizures, the defendant “does not point out any reasons why we should depart from the analysis that the United States Supreme Court has developed in its decisions.” Id. at 339. In Greene, the court in fact stated what both the state and defendant implicitly agreed in the trial court in the present case, that Article I, section 9, provides “at least as much protection” as the Fourth Amendment. Id.

In short, as discussed above, the terms of the state’s assignment of error are broad enough to encompass all bases for the trial court’s decision. Further, the state supports its assignment of error in regard to probable cause and exigent circumstances with citations to cases decided under both Article I, section 9, and the Fourth Amendment. Finally, none of the cases relied on by either party suggests that the analysis under the Fourth Amendment could result in suppression under that provision but not under Article I, section 9.31 would conclude that the state has adequately challenged on appeal the trial court’s rulings on probable cause and exigent circumstances.

*425Turning to the merits, I would hold that the trial court erred in granting defendant’s motion to suppress because, in my view, the search was supported by probable cause and exigent circumstances. Under Article I, section 9, of the Oregon Constitution and the Fourth Amendment to the United States Constitution, warrantless searches are considered unreasonable unless the state shows, by a preponderance of the evidence, that a search falls within an established exception to the warrant requirement, such as probable cause accompanied by exigent circumstances. State v. Davis, 295 Or 227, 237, 666 P2d 802 (1983). In general, under Article I, section 9, probable cause exists when an officer subjectively believes that a crime has been committed and that belief is objectively reasonable under the circumstances. State v. Cardell, 180 Or App 104, 110, 41 P3d 1111 (2002) (citing State v. Owens, 302 Or 196, 204, 729 P2d 524 (1986)). In determining whether objective probable cause exists, we look to the totality of the circumstances and the reasonable inferences that may be drawn from those circumstances, but no single factor necessarily is dispositive. State v. Spruill, 151 Or App 87, 90-91, 948 P2d 726 (1997). Whether objective probable cause exists is a matter of law. State v. Kappel, 190 Or App 400, 404, 79 P3d 368 (2003), rev den, 336 Or 509 (2004). Exigent circumstances exist if a situation requires “the police to act swiftly to prevent danger to life or serious damage to property, or to forestall a suspect’s escape or the destruction of evidence.” Stevens, 311 Or at 126.

The following circumstances of the present case are relevant to the determination that the officer’s subjective belief that a crime had been committed was objectively reasonable: (1) Defendant, who lived alone, left his home to play in a televised basketball game in a nearby city several hours before the burglar alarm at his house went off; (2) neighbors saw a car that they did not recognize leave defendant’s driveway after defendant left home but before the burglar alarm sounded; (3) when the police arrived to investigate the alarm, the front door to defendant’s home was open; and (4) when the police arrived to investigate the alarm, it was after dark on a winter evening.

As an initial matter, I disagree with the trial court’s conclusion that the sounding of the burglar alarm added *426nothing to the probable cause equation. Although the trial court may well be correct that the sounding of a burglar alarm alone does not establish probable cause, it is certainly an overstatement to say that, under the totality of the circumstances, the alarm contributes nothing to an assessment of probable cause. Residential burglar alarm systems are designed to detect unauthorized entries into residences. The fact that such systems often sound in the absence of unauthorized entries into residences in no way suggests that they are completely useless in detecting unauthorized entries into residences. Defendant himself continued to use the alarm system despite the many false alarms at his house.

In the present case, the sounding of the burglar alarm is of particular significance based on all of the other circumstances.4 Defendant, the sole occupant of the house, was known to have left the house to play in a basketball game several hours before the alarm sounded and, thus, could not have triggered the alarm. That fact, added to the facts that an unfamiliar car was seen pulling out of the property after defendant had left the house and before the alarm went off and that the front door was found open on a winter evening, adds up to objective probable cause.

This case differs from State v. Bramson, 94 Or App 374, 765 P2d 824 (1988), on which defendant relies, in several notable respects. In Bramson, an unnamed source reported that the front door was open at a residence and that its residents were out of town. Officers arrived and saw that there was a vehicle in the driveway and that the door of the house was open. The screen door was closed but had no screen in the center section. Id. at 376. The officers entered the residence and discovered drugs in the house. We concluded that the officers lacked probable cause to enter the house:

*427“The officers knew only that the front door was open and that the screen door had no glass or screen, that the weather was cold, that a vehicle was in the driveway and that the residents were reportedly out of town. There was, however, no report or other indication of unauthorized entry or criminal activity. The broken door screen and the open front door would not have led a prudent officer to believe that a burglary was being or had been committed. State v. Apodaca, 85 Or App 128, 133, 735 P2d 1264 (1987). The open door was at least as consistent with the occupants’ neglect as it was with a burglary.”

Id. at 377.

Although the present case has some factual similarities — cold weather, an open door, and occupants known not to be at home — it has an important difference. In Bramson, there was no way to know if the occupants had left the door open before they left town. As noted, the open door there was as consistent with occupant neglect as it was with burglary. Id. Here, in contrast, a burglar alarm sounded, providing information on the timing of the possible entry into the house. It is a reasonable inference that the opening of defendant’s front door likely triggered the burglar alarm and that that event occurred at a time when the neighbors knew that defendant, the sole occupant, was not at home. Under those circumstances, in contrast to Bramson, neither the open door nor the sounding of the burglar alarm was consistent with occupant neglect.

The present case also bears little similarity to Apodaca, in which officers investigating an abandoned car went to an address on an envelope they had found in the car. 85 Or App at 130. They discovered that the front door screen was broken and that the front door was open. They knocked and received no response, then entered. Id. The state argued that the police had reason to believe that a burglary was being or had been committed. Id. at 132. We rejected that argument, noting that “a broken front door screen and an open front door would not lead a prudent and reasonable officer to believe that a burglary was being committed or had been committed.” Id. at 133. Compare State v. Christenson, 181 Or App 345, 45 P3d 511 (2002) (open door and report by neighbor that dogs were roaming at large did not provide *428probable cause that burglary was in progress), with Collier v. City of Portland, 57 Or App 341, 644 P2d 1139 (1982) (officer had probable cause to enter after observing open, broken door, neighbor had reported noise, and items including a television and radio were on the porch).

Judge Armstrong’s concurrence reasons that the door must have been opened from the outside, which triggered the alarm. 198 Or App at 410 (Armstrong, J., concurring). The concurrence posits that the car that was seen leaving defendant’s driveway shortly before the burglar alarm sounded could have had no causal relationship to the sounding of the alarm, apparently based on the unsupported assumption that the car could only transport people away from defendant’s residence and not to defendant’s residence. Id. at 410.

Judge Armstrong’s speculations are without any support in the record. Although he declares it “apparent,” there is no evidence in the record that tells us, or would have told the officers, whether the door had been opened from the inside or from the outside. He also jumps to the conclusion that it is more likely here that what happened is that defendant accidentally left the door ajar when he left and the door somehow opened on its own. Id. at 410. Even if I accept the concurrence’s apparent assumption that an alarm would be set on a door that was left ajar, there is no evidence offering any explanation for how the door could have opened on its own. It seems far more likely that the alarm went off because the door was opened by a person. The concurrence also speculates that, because a neighbor saw the strange car at the house drive away 45 minutes before the alarm went off, “the person or people in the car presumably did not cause the alarm to sound by opening the door.” Id. at 410 (Armstrong, J., concurring). How can we presume that? We have no idea how many people were in the car or what they were doing at the house or whether all of the occupants of the car left or someone was dropped off. We know only that it was a car that the neighbor did not recognize and that it was at the house not too long before the alarm went off. Further, the concurrence’s conclusion that no one was in the house because the sounding of the alarm would have scared off any burglar again is nothing more than speculation. Finally, the fact that *429the neighbors did not see anyone in or around the house did not tell the police anything. Neighbors watching a house from a distance cannot see all of the outside or the inside of a house. The police had no way of knowing, short of going in, whether an intruder was in the house.

Bramson and Apodaca demonstrate that an open door is not enough to establish probable cause that a burglary is underway. The present case, however, is not simply an “open door” case. Under the circumstances of this case, the officers reasonably inferred that the opening of the door triggered the burglar alarm that had been set to detect unauthorized intruders, and that the sole occupant was not at home at the time that the alarm went off and, therefore, could not have triggered the alarm himself. As noted above, alarms are often triggered when homeowners set them off while leaving home. That did not happen here; the alarm sounded long after defendant left the house. In addition to the alarm going off and the door being open, the police knew that, here, a strange car had been in the driveway earlier. Based on all of the above circumstances, it was probable that an intruder had entered defendant’s house. Accordingly, I would conclude that the officers had objective probable cause to believe that an intruder had activated the alarm.

I would reach the same conclusion under the Fourth Amendment. The case law from other jurisdictions interpreting the Fourth Amendment uses a similar analysis. See, e.g., A.A.G. v. State, 668 So 2d 122 (Ala Crim App 1995) (probable cause supported entry into house where burglar alarm had sounded, where there were no indications of a break-in, and teen inside the house first turned off light and did not answer door, then opened door and explained she lived there); State v. Smith, 49 NC App 293, 271 SE2d 86 (1980) (officer had probable cause to believe crime had been committed where burglar alarm at deserted store was activated, a person had been seen running nearby, and defendant was in a telephone booth 40 feet from the store); State v. Proctor, 12 Wash App 274, 529 P2d 472 (1974), rev den, 85 Wash 2d 1010 (1975) (officer who responded to alarm at real estate office and entered through open door lawfully observed stolen property); State v. DiGiallonardo, 160 Mont 379, 503 P2d 43 (1972) (officers had probable cause to believe crime had been *430committed or was being committed where store alarm sounded and men nearby began to rim when they saw officers); Sipera v. State, 286 Minn 536, 175 NW2d 510 (1970) (same); People v. Williams, 67 Cal 2d 226, 60 Cal Rptr 472, 430 P2d 30 (1967) (officer had probable cause to arrest defendant who was seen driving a block and a half away from a store whose alarm had sounded, and no other moving cars were in the area).

The question remains, however, whether exigent circumstances permitted the officers’ warrantless entry into the house to investigate the circumstances of the intrusion. As noted above, exigent circumstances are those that require “the police to act swiftly to prevent danger to life or serious damage to property, or to forestall a suspect’s escape or the destruction of evidence.” Stevens, 311 Or at 126.

The question here is whether there was an imminent threat to defendant’s property or the potential that a burglar inside the /house could have escaped before a search warrant was obtained. The possibility5 also existed that, if the officers did not search the residence for intruders, the occupant could return home and be confronted by an intruder.6 That is not a possibility that one would expect police officers responding to a call concerning a potential break-in to ignore. Given that there was probable cause to believe that an intruder had activated the alarm, could a responding officer be expected to leave the scene in order to secure a search warrant? Even if two officers were present, as was the case here, must an officer be left without backup to watch all exits from a residence while another officer leaves to obtain a warrant? Alternatively, more officers could have been called to surround the residence while a search warrant *431was obtained in person or by telephone, in order to prevent any intruder from leaving and to prevent the occupant from entering. Realistically, though, such measures would not ensure the protection of the occupant’s property from the intruder, nor would they be as effective at preventing escape as would taking immediate steps to locate and apprehend any intruder. Judge Armstrong’s concurrence would have the officers simply walk away. This court has not previously second-guessed officers in such a case by imposing such restraints on them in responding to circumstances such as these, and I would not do so here.

Defendant argues that facts that merely suggest a completed burglary do not give rise to exigent circumstances. However, the facts here do not suggest a completed burglary. To recapitulate the relevant facts, the alarm sounded at approximately 6:30 p.m., shortly after an unfamiliar car was seen leaving the driveway to the house. The first officer arrived some 15 minutes later. One of the neighbors who reported the alarm had watched the house from his nearby house until the police arrived and did not report that he had seen anyone leaving from his vantage point. Thus, while the facts indicated that an intruder may have set off the alarm by opening the front door, no facts supported a belief that such an intruder had left the premises. Although it is true that a strange car was seen leaving defendant’s house that evening, the car was seen leaving before the burglar alarm was heard. Thus, although the car added to the suspicious circumstances indicating that an intruder may have entered defendant’s house, the fact that the car had left defendant’s house before the burglar alarm was heard does not support an inference that whoever had set off the burglar alarm must necessarily have left defendant’s house.

Given all of the above circumstances, I would conclude that exigent circumstances existed because of the potential that an intruder remained inside the house. See, e.g., State v. Lynch, 135 Or App 528, 533, 900 P2d 1042, rev den, 322 Or 362 (1995) (where probable cause existed that a burglary was in progress, exigent circumstances justified entry and search into areas that could conceal a burglar); Collier, 57 Or App at 346 (“Having probable cause to believe *432that a burglary was being committed and that the perpetrator was still on the premises, the police were privileged to enter[.]”); State v. Schrag, 21 Or App 655, 657, 536 P2d 461 (1975) (“circumstances known to the officers would warrant a prudent person in believing that a felony was being committed, and in concluding that prompt entry was necessary to apprehend the suspects”); United States v. Johnson, 9 F3d 506, 508 (6th Cir 1993), cert den, 512 US 1212 (1994) (police may enter a residence without a warrant if they have probable cause to believe a burglar is inside); United States v. Singer, 687 F2d 1135, 1144 (1982), on reh’g en banc, 710 F2d 431 (8th Cir 1983) (same); State v. Londo, 252 Wis 2d 731, 643 NW2d 869 (Wis App), rev den, 254 Wis 2d 263 (2002) (same).

In my opinion, the trial court erred in granting defendant’s motion to suppress evidence obtained as a result of the officers’ entry into defendant’s house. Accordingly, I believe that this case should be reversed and remanded, and I dissent.

Edmonds, Haselton, and Linder, JJ., and Leeson, J. pro tempore, join in this dissent.

dissenting.

For the reasons stated in their opinions, I agree with Judge Deits and Judge Haselton that the state’s assignment of error is broad enough to encompass all grounds ruled on by the trial court and that the trial court erred in suppressing the evidence seized from defendant’s residence. I write separately, however, to express my view that Judge Armstrong errs in his concurring opinion by considering circumstances not known to the officers at the time of their entry into defendant’s residence for purposes of determining whether they had probable cause and exigent circumstances to search.

Judge Armstrong, reciting from the trial court’s findings, says that there had been 32 false alarms at defendant’s residence over a six-year period, that defendant had been cited approximately seven times under a Lake Oswego false alarm ordinance, and that, during the time period that the search occurred, over 99 percent of all residential alarms *433received by the city police were false alarms. 198 Or App at 402 (Armstrong, J., concurring). Indeed, the trial court found:

“During the one-year period from July 2001 to July 2002, 2,026 alarms occurred that came to the attention of the Lake Oswego Police Department. Of those alarms, 1,327 were residential. Of the residential alarms, all but two were false alarms. Expressed as a percentage, 99.85 percent of all residential alarms were false alarms [.]”

The trial court concluded that “[t]he false alarm rate in Lake Oswego is so high that a sounding alarm cannot add to the showing of probable cause [.]” The trial court erred, and Judge Armstrong errs, by including those facts in the calculus to determine probable cause when that information was unknown to the officers at the time that they entered defendant’s residence.

The test for determinating probable cause to search must be based not on what the officer could believe but on the information that the officer “actually believed, based upon the underlying facts of which he was cognizant, together with his own training and experience.” State v. Owens, 302 Or 196, 204, 729 P2d 524 (1986); see also State v. Groda, 285 Or 321, 324, 591 P2d 1354 (1979) (“[t]he searching officer personally must have information which constitutes probable cause, or the searching officer must be directed to make the search by an officer who personally has that knowledge”).

In this case, the officers who conducted the search were Brady and Forman. Brady testified, in part:

“Q. Were you aware, Officer Brady, that from July 2001 to July 2002, you, individually, responded to 42 false alarms in the course of your duties as a police officer in Lake Oswego?

“A. I did not know that.

* % * *

“Q. Over the years that you worked there, can you give us an estimate of how many false alarms you probably responded to?

“A. I couldn’t even hazard to guess.

“Q. Hundreds?

*434“A. Probably, if what you are saying is true, 42 a year, if that’s an average, times seven.”

Officer Forman testified, in part:

“Q. * * * You are familiar, are you not, at least generally, with the high incidence of false alarms in Lake Oswego?

<t% $ ‡ ‡ ‡

“A. Oh, high number. Yes.

“Q. Are you familiar with just how high a percentage of false alarms there are?

“A. No. All I could say about that is there are many. I don’t know percentages.

“Q. Would you be surprised to know that for the last reporting year, from July 2001 to July of this year, its over 99 percent false alarms?

“A. No, that wouldn’t surprise me. Like I said, I don’t know the percentages, but many false alarms.”

Later, Forman was asked whether he was familiar with the age of the alarm system in defendant’s residence and whether he was aware of repairs to the alarm because of malfunctions. Forman replied, “I have no idea what might or might not have been done with [defendant’s] alarm. I don’t know.”

The first component of the test for probable cause under Owens is to determine the facts that the officers knew when they arrived at defendant’s residence after the police received a phone call from a neighbor. Forman, the supervising officer, summarized those facts in his testimony at the suppression hearing:

“Well, I arrived and talked to Officer Brady, and I knew there had been an alarm at the residence, that the door, front door, was open, and that there had been a car that had left sometime prior to the alarm. That from my understanding, it wouldn’t have been [defendant], who I understood resided there, but somebody else. It was my understanding that [defendant] had left a couple of hours before, or sometime prior, and that there was — supposedly been another car seen leaving after that.”

*435Significantly, neither officer testified that he was aware of the number of false alarms that had occurred at defendant’s house or the percentage of false alarm reports made during the previous year to the Lake Oswego Police Department at the time that they entered the residence.

The second component of the test from Owens for purposes of determining probable cause is the officers’ training. In that regard, Forman testified that, when a burglar alarm goes off at a residence and no one is found at the residence, the officers are trained to

“check windows and doors and make sure it’s all secure. If everything is secure, all appears to be in order, there isn’t anything out of the ordinary, * * * anything other than you have an alarm and no visible entry or the building not— isn’t open. You just leave a false alarm notice at the location and leave.”

When, however, the premises are unsecured, such as having been left with an open door, officers are trained “to go ahead and make entry into the premises and check for persons.” Besides checking to determine if there are persons inside the residence, officers are also trained to investigate whether there are

“[s]igns of a crime that could have been committed; say, a ransacked house. Depending what room you might be in, say, in the living room where there might be sound equipment, TVs, VCRs, where things are piled up on the floor, does it look like things are missing? Sometimes, if there had been an entry into a home and an alarm went off, say, for example, somebody had been frightened off and left, there could be items piled up as if someone was going to take them out. So you are looking for people, for signs that things don’t appear as they probably should be in somebody’s home.”

Finally, the third component of probable cause under the Owens test is the officers’ experience. Brady testified that, in his experience, a common method of entry for burglars is through an unlocked door. The combination of the circumstances of a burglar alarm and an open door when the resident is not present constitutes a situation where, based on his experience, Brady “attempt[s] to make sure that *436nobody has entered the residence and might be inside.”1 As to the officers’ experience or knowledge of the incidence of false alarms at defendant’s residence and throughout the city, this record demonstrates that neither officer was aware of the number of false alarms at defendant’s residence or the percentage of false alarms received by the Lake Oswego Police Department at the time of their entry into defendant’s residence. Also, Brady was unaware of the number of false alarms to which he had previously responded. He commented, “I couldn’t even hazard a guess.” Forman testified that he was not familiar with “how high a percentage of false alarms” are received by the police department, although he agreed with defense counsel that there are “many.”2 Forman also was asked about his experience regarding the incidence of false alarms at defendant’s residence. He said, “I have no idea[,] * * * I don’t know.” Even when defendant is given the benefit of all reasonable inferences that can be drawn from the officers’ testimony, the most that can be said from the record before us is that the officers were aware that “many” alarms that they respond to turn out to be false.

“Probable cause” to conduct a warrantless search does not require certainty. State v. Collicott, 56 Or App 605, 608, 642 P2d 1187, rev den, 293 Or 190 (1982). In general, probable cause to search exists when a reasonable person would believe it is more likely than not that the objects of a search will be found at the location to be searched. State v. Anspach, 298 Or 375, 380, 692 P2d 602 (1984). Here, the object of the officers’ search was to determine if the crime of burglary had been or was being committed at the time the officers arrived. The issue of the reasonableness of the search turns on what information the officers were cognizant of at that time in light of their training and experience. In my view, even though the officers knew that “many” false alarms are reported to the Lake Oswego Police Department *437annually, a reasonable person would believe it more likely than not that defendant’s residence had been or was the subject of an ongoing burglary when the officers arrived. That is because the officers learned that defendant had left the residence several hours earlier, that a burglar alarm had gone off after he left without any apparent explanation for its occurrence, that another car had been observed leaving the residence after that time, and that the front door was left open. Those facts permit a reasonable belief that a burglary had occurred, or was occurring, and they outweigh the officers’ knowledge that many alarms in the city ultimately turn out to be false.

In his, concurring opinion, Judge Armstrong responds to the above analysis by reasoning that the state does not argue that the information not known to the officers could not be considered and that, although “[t]hey did not know that the rate of false alarms was 99 percent, * * * they knew that a sounding alarm did not have much if any bearing on whether a burglary had occurred, which is the conclusion that the trial court reached.” 198 Or App at 409 n 6 (Armstrong, J., concurring). The concurrence is wrong on both counts.

ORS 138.220 provides the standard of review that is applicable to the order in this case. It provides that the order “can be reviewed only as to questions of law appearing upon the record.” With regard to Judge Armstrong’s assertion that the officers “knew that a sounding alarm did not have much if any bearing on whether a burglary had occurred,” the officers did not testify to that fact, as the above quotations of their testimony illustrates. Although hypothetically a fact-finder could draw such an inference about the officers’ mental state based on their experiences, the trial court in this case did not make such a finding. Rather, its findings in its order about the rate of false alarms are made with regard to the issue of “objective probable cause,” which is solely a question of law. This court’s “function is to decide whether the trial court applied legal principles correctly to [the] facts” that it did find and that are supported by the evidence. State v. Stevens, 311 Or 119, 126, 806 P2d 92 (1991). By considering information not known to the officers at the time that they made the entry into defendant’s residence, the trial court *438committed an error of law for the reasons described above. It is certainly within our authorized scope of review to correct that error by considering only those facts that are legally relevant to the legal question of objective probable cause.

Further, Judge Armstrong’s assertions that the state does not make the argument that the trial court erred by considering information not known by the officers at the time of entry and that this court therefore should not address the trial court’s error in that respect is puzzling and proceeds from a premise that is, with respect, not completely factually accurate. There is no question of preservation under ORAP 5.45 regarding the issue of objective probable cause in this case. Both parties argued the issue to the trial court, and the trial court expressly ruled on that issue. Moreover, the state expressly argues on appeal that the trial court erred in deciding that there was not objective probable cause under the totality of the circumstances: “In making the determination of objective reasonableness, this court examines the totality of the facts and circumstances and reviews anew the trial court’s legal conclusions.” Although it is correct that the state on appeal never expressly asks this court to limit its analysis of objective probable cause to the information known to the officers at the time of entry, the state does argue the import of the false alarm evidence as it comments on the issues of objective probable cause and exigent circumstances. For example, the state asserts, “Although it was possible that the alarm was a false one - especially in light of the large number of false alarms in Lake Oswego - the totality of the circumstances suggested otherwise.”

The preceding analysis of objective probable cause based on the facts of which the officers were cognizant, together with their training and experience, is based on the “totality of the circumstances,” as both the state and defendant correctly urge. But not all the circumstances in this case are germane to the determination of objective probable cause as the previously cited case law indicates. Contrary to the implication of Judge Armstrong’s assertion, I am unaware of any cognizable legal principle of appellate review that prevents this court from correctly applying a rule of law by excluding evidence that is not relevant to the determination of a legal question merely because the parties do not make a *439particular argument about that evidence. Rather, I submit that this court has an affirmative duty to apply the law correctly when confronted with a question of law. Otherwise, we perpetuate an incorrect understanding of the law.

Accordingly, I dissent.3

dissenting.

I join in Judge Deits’s analysis of the lawfulness of the search but write separately to supplement the discussion of the threshold procedural question.

So, what is really going on here — procedurally?

Defendant moved to suppress, arguing that the search violated both Article I, section 9, of the Oregon Constitution, and the Fourth Amendment to the United States Constitution. If defendant prevailed on either of those grounds, he was entitled to suppression. Nevertheless, under State v. Kennedy, 295 Or 260, 262, 66 P2d 1316 (1983), the trial court was obligated to address the state constitutional argument first — and could properly reach the merits of defendant’s federal constitutional challenge only if it rejected defendant’s Oregon law-based grounds for suppression. Id.; see also Zockert v. Fanning, 310 Or 514, 520, 800 P2d 773 (1990).

The state argued that defendant was not entitled to suppression under either the Oregon Constitution or the Fourth Amendment.

The trial court granted the motion to suppress, concluding that the search was unlawful under the Oregon Constitution. The trial court also concluded that the search was unlawful under the Fourth Amendment.

*440The state appealed, assigning error to the allowance of the motion to suppress. The state briefed the lawfulness of the search under Article I, section 9, but presented no separate argument concerning the lawfulness of the search under the Fourth Amendment.

Defendant’s responding brief addressed only the propriety of suppression under the Oregon Constitution. Defendant’s brief did not assert that the trial court’s suppression ruling should be affirmed on the alternative federal constitutional ground.

At oral argument, neither counsel — both of whom are skilled and extremely experienced — mentioned the Fourth Amendment. Neither did we.

It was only after oral argument that we, for the first time, tumbled to the fact that the state had not expressly challenged the trial court’s alternative Fourth Amendment-based ground for suppression — and that defendant had not invoked the Fourth Amendment as an alternative basis for affirmance.

According to the lead opinion and Judge Landau’s concurrence, all of this leaves us in a posture in which we must affirm the order of suppression, regardless of whether the search was, in fact, lawful under both the Oregon Constitution and the United States Constitution. We are to do so notwithstanding that (a) the lawfulness of the search under the Fourth Amendment was fully preserved below; (b) the state properly assigned error to the trial court’s ruling granting the motion to suppress; and (c) defendant never substantively invoked the Fourth Amendment in his respondent’s brief or at oral argument. While I agree with Judge Landau that “[s]auce for the goose, [is] sauce for the gander,” see 198 Or App at 416 (Landau, J., concurring), the more tempting sound bite here is “a plague on both their houses.”

Still, it is a “plague” that we must confront and resolve. And, with respect, I believe that Judge Deits’s resolution of the procedural conundrum in this contorted case does not somehow unfairly differentiate between criminal defendant “geese” and the state “gander.” Rather, that resolution accords with two neutral principles:

*441First, like it or not, under Kennedy, the trial court’s “holding” on the Fourth Amendment ground was procedurally improper — or, more accurately, not a “holding” at all. Although the trial court’s determination of both the state and federal constitutional challenges might well be understandable as a practical matter,1 it violated Kennedy's “first things first” precept. See Zockert, 310 Or at 520 (“This court decides cases upon subconstitutional grounds, where available, even though litigants argue only constitutional errors. Likewise, the state constitution is consulted before the federal.”); see also State v. O’Donnell, 192 Or App 234, 239 n 5, 85 P3d 323 (2004) (trial court’s resolution of constitutional questions after determination of case on statutory grounds was “procedurally erroneous”).2

It is true, as Judge Landau points out in his concurring opinion,3 that nothing in Kennedy expressly prohibits an Oregon court from reaching and resolving federal constitutional issues even if the court’s determination under state law is dispositive. But we have always so understood and applied Kennedy. And so has the Oregon Supreme Court. Since Kennedy, neither court has ever proceeded to address a federal constitutional question after having decided the appeal on state-law grounds.

Thus, the trial court’s determination with respect to the Fourth Amendment was, at best, a contingent observation — more or less the obverse of what we do, on occasion, with respect to matters that we consider likely to arise on *442remand. The trial court’s “holding” under the Fourth Amendment was, in effect, a statement that, “if the Court of Appeals disagrees with my ruling under the Oregon Constitution, don’t send this back to me to consider defendant’s Fourth Amendment challenge because here’s how I would rule on that.”

I emphasize that this is not a situation, like many that we commonly encounter, in which the trial court has granted a dispositive motion or entered judgment on several independently sufficient grounds — e.g., allowing a defense motion for summary judgment in a contract case on both statute of limitations and statute of fraud grounds, or entering a judgment of termination of parental rights based, alternatively, on findings of unfitness, ORS 419B.504, and neglect, ORS 419B.506. The difference lies in Kennedy, which bars consideration of a federal constitutional matter if the state constitutional ground is independently dispositive.

Second, given the foregoing, it was not incumbent on the state, as appellant, to take issue with the trial court’s federal constitutional “holding” — because, as a matter of Oregon jurisprudence, that “holding” was not, and could not be, an independently sufficient ground for the trial court’s ruling. Rather, it was up to defendant, if he so chose, to argue to us that, if we disagreed with the trial court and concluded that the search was not unlawful under the Oregon Constitution, the Fourth Amendment nevertheless afforded an alternative ground for affirmance. If defendant had done so, the state would, no doubt, have responded either with a reply brief, at oral argument, or both. But that, of course, did not happen.

Thus, if the trial court and defendant here had acted in a manner consonant with Kennedy, we would have been in a position of either (a) ourselves addressing the merits of the alternative Fourth Amendment ground for suppression on the evidentiary record developed before the trial corut; or (b) remanding for the trial corut to resolve that matter. In other words, we would have been, and should have been, in the same procedural posture from which Judge Deits’s dissent now operates.

I agree with Judge Deits’s analysis of the lawfulness of the search under both Article I, section 9, of the Oregon *443Constitution, and the Fourth Amendment to the United States Constitution. Accordingly, I dissent.