State v. Dean

HURWITZ, Justice. ¶ 1 We granted review in this case to address a recurring Fourth Amendment issue — whether the search of an automobile is “incident” to the arrest of a defendant. In this case, the search occurred after the defendant was arrested in a house, some two and one-half hours after he had exited the vehicle. The superior court held that the search violated the Fourth Amendment; the court of appeals reversed. For the reasons below, we hold that the superior court correctly found that the search in this case did not fall within the “search incident to arrest” exception to the Fourth Amendment’s warrant requirement. I. ¶ 2 On February 7, 2001, Phoenix police officers received a tip that Donald Dean, the subject of two felony drug arrest warrants, was residing at a house on East Cholla Street in Phoenix. The police set up surveillance of the house and, at about 6:00 p.m., observed a Jeep Grand Cherokee leaving the residence, driven by a person fitting Dean’s description. Several officers followed the Jeep in a marked patrol car; after a short time, the officers activated their overhead lights. Dean, who was in fact driving the Jeep, did not pull over, but instead returned to the East Cholla residence and parked in the driveway. He jumped out of the Jeep, leaving the keys in the ignition, and ran into the garage of the house. One of the officers then took the keys from the Jeep. ¶ 3 The officers on the scene summoned a tactical team from the Phoenix Special Assignment Unit. After obtaining permission from the owner of the house, the tactical team went inside. The team eventually found Dean hiding in the attic, arrested him, and took him to a waiting patrol car. ¶ 4 Dean’s arrest occurred approximately two and one-half hours after he fled the Jeep. After Dean was arrested, officers searched the Jeep and discovered methamphetamine in the passenger compartment. Based on this discovery, the police obtained a warrant to search the residence and, in executing that warrant, discovered additional quantities of methamphetamine, marijuana, drug paraphernalia, and weapons. Dean was subsequently charged with possession of equipment or chemicals for the manufacture of dangerous drugs, possession of dangerous drugs for sale, and possession of drug paraphernalia. ¶ 5 Dean filed a motion in the superior court to suppress all evidence seized from the Jeep, alleging unlawful search and seizure. The superior court granted the motion to suppress. The trial court rejected the State’s contentions that the vehicle was abandoned and that the search was simply an administrative inventory of the vehicle contents. The superior court also rejected the State’s argument that the warrantless search of the Jeep was incident to Dean’s arrest, noting that “[t]he arrest took place two and a half hours later at a different location.” ¶ 6 The State timely appealed the suppression order pursuant to Arizona Revised Statutes (“A.R.S.”) § 13-4032(6) (2001). The court of appeals reversed, holding that the *161 Jeep search fell within the “incident to arrest” exception to the Fourth Amendment’s warrant requirement. State v. Dean, 203 Ariz. 408, 409 ¶ 1, 55 P.3d 102, 103 (App. 2002). The court of appeals reasoned that because the police could have searched the vehicle incident to an arrest if Dean had been apprehended either inside the vehicle or directly outside of it, Dean could not “evade a search by leaving the vehicle before the officers arrest him.” Id. at 411 ¶ 10, 55 P.3d at 105. ¶ 7 Dean filed a petition for review, and we granted review to address the applicability of the “incident to arrest” exception to the warrant requirement in this situation. We have jurisdiction pursuant to Article 6, Section 5(3) of the Arizona Constitution, Arizona Rule of Criminal Procedure 31.19(c)(3), and A.R.S. §§ 13-032(6) and -4033(A)(2) (2001). II. ¶ 8 The Fourth Amendment to the United States Constitution, made applicable to the States through the Due Process Clause of the Fourteenth Amendment, protects against “unreasonable searches and seizures,” and provides that search warrants shall be issued only upon “probable cause.” U.S. Const. amend. IV. Searches conducted without a judicially approved warrant “are per se unreasonable under the Fourth Amendment — subject only to a few specifically established and well-delineated exceptions.” Katz v. United States, 389 U.S. 347, 357, 88 S.Ct. 507, 19 L.Ed.2d 576 (1967) (internal citations omitted). The State suggests that the warrantless search of Dean’s Jeep can be justified by virtue of three of those “exceptions”: (1) because the Jeep was “abandoned”; (2) because the search was an administrative “inventory” of its contents; and (3) because the search was “incident” to Dean’s arrest. 1 A. ¶ 9 The State’s first two arguments do not require extended discussion. The superior court specifically found that the Jeep, which was parked in the driveway of Dean’s residence, was not abandoned. In reviewing an order involving a motion to suppress, we review the facts in the light most favorable to sustaining the order, and will not disturb the trial court’s ruling absent “clear and manifest error.” State v. Hyde, 186 Ariz. 252, 265, 921 P.2d 655, 668 (1996). See State v. Jones, 203 Ariz. 1, 5 ¶ 8, 49 P.3d 273, 277 (2002) (“Clear and manifest error ... is really shorthand for abuse of discretion.”). We find no such error here. ¶ 10 The State’s attempt to justify the search as an “inventory” of the Jeep fails on similar grounds. While police have the power to perform a warrantless “administrative” search of an impounded vehicle for “community caretaking functions,” see South Dakota v. Opperman, 428 U.S. 364, 368, 96 S.Ct. 3092, 49 L.Ed.2d 1000 (1976), such a search must be “routine,” and not “a pretext concealing an investigatory police motive.” Id. at 376, 96 S.Ct. 3092. The officer who conducted the search of the Jeep testified at the suppression hearing that his purpose was “to search for evidence.” In light of that testimony, the superior court did not err in concluding that the search was not an administrative inventory. 2 B. ¶ 11 We therefore turn to the only remaining applicable justification for the warrantless search here, the “search incident to arrest” exception. In Chimel v. Califor *162 nia, 395 U.S. 752, 89 S.Ct. 2034, 23 L.Ed.2d 685 (1969), the Supreme Court explained both this exception and its justifications. Chimel involved the arrest of a defendant inside his home and the subsequent warrant-less search of the home. The California Supreme Court upheld the search as “incident to a valid arrest.” Id. at 755, 89 S.Ct. 2034. In an opinion by Justice Stewart, the Supreme Court of the United States reversed. ¶ 12 Chimel began from the premise that when a search is conducted without a warrant, “[t]he scope of [a] search must be ‘strictly tied to and justified by’ the circumstances which rendered its initiation permissible.” Id. at 762, 89 S.Ct. 2034 (alteration in original) (quoting Terry v. Ohio, 392 U.S. 1, 19, 88 S.Ct. 1868, 20 L.Ed.2d 889 (1968)). When a search is conducted incident to a valid arrest, Justice Stewart explained, two sets of circumstances justify departure from the general warrant requirement: When an arrest is made, it is reasonable for the arresting officer to search the person arrested in order to remove any weapons that the latter might seek to use in order to resist arrest or effect his escape. Otherwise, the officer’s safety might well be endangered, and the arrest itself frustrated. In addition, it is entirely reasonable for the arresting officer to search for and seize any evidence on the arrestee’s person in order to prevent its concealment or destruction. And the area into which an arrestee might reach in order to grab a weapon or evidentiary items must, of course, be governed by a like rule. Id. at 763, 89 S.Ct. 2034. ¶ 13 Chimel taught that these twin aims of the search incident to arrest exception — officer safety and preservation of evidence— provide “ample justification” for a warrant-less search “of the arrestee’s person and the area “within his immediate control’ — construing that phrase to mean the area from within which he might gain possession of a weapon or destructible evidence.” Id. The search in Chimel was of the defendant’s entire house, taking between forty-five minutes and an hour after the arrest. Because the search “went far beyond the petitioner’s person and the area from within which he might have obtained either a weapon or something that could be used as evidence against him,” the Court held that there was no constitutional justification for departing from the general warrant requirement and that the search was unreasonable under the Fourth Amendment. Id. at 768, 89 S.Ct. 2034. C. ¶ 14 Under the rule announced in Chimel, determining whether a particular area in which evidence was found was within an arrestee’s “immediate control” required an examination of the facts and circumstances surrounding each arrest. This case-by-case analysis “presented a significant burden to courts and police” when the arrest occurred in or near an automobile and police had to decide at the time of the arrest which portions of the automobile were within the arrestee’s immediate control. Glasco v. Commonwealth, 257 Va. 433, 513 S.E.2d 137, 143 (1999) (Lacy, J., concurring). In New York v. Belton, 453 U.S. 454, 101 S.Ct. 2860, 69 L.Ed.2d 768 (1981), the Supreme Court recognized the problem, noting that the “courts have found no workable definition of the ‘area within the immediate control of the arrestee’ when that area arguably includes the interior of an automobile and the arrestee is a recent occupant.” Id. at 460, 101 S.Ct. 2860 (quoting Chimel, 395 U.S. at 763, 89 S.Ct. 2034). Because a “‘single familial’ standard is essential to guide police officers,’ ” id. at 458, 101 S.Ct. 2860 (quoting Dunaway v. New York, 442 U.S. 200, 213-14, 99 S.Ct. 2248, 60 L.Ed.2d 824 (1979)), the Court undertook in Belton to establish such a “workable rule.” Id. at 460, 101 S.Ct. 2860. ¶ 15 Belton arose from the stop of a speeding vehicle on the New York State Thruway. The police officer stopped the vehicle and, after examining the registration, discovered that none of the four men inside owned the vehicle or was related to its owner. The officer also smelled burnt marijuana and saw an envelope marked “Supergold” on the floor of the vehicle. He then removed the four individuals from the vehicle, separated them, and searched the vehicle. In the back seat he found a leather jacket belonging to Belton; inside a zipped pocket, the officer dis *163 covered cocaine. Id. at 455-56, 101 S.Ct. 2860. ¶ 16 The New York Court of Appeals held that because there was no danger that the arrestee or his confederates could gain access to the “zippered pockets of an unaceessible jacket,” id. at 456, 101 S.Ct. 2860, the search exceeded the scope justified by the twin goals of the Chimel exception and was not validly incident to Beltons arrest. The Supreme Court reversed. Once again writing for the Court, Justice Stewart noted that “articles inside the passenger compartment of an automobile are in fact generally, even if not inevitably, within ‘the area into which an arrestee might reach in order to grab a weapon or evidentiary ite[m].’ ” Id. at 460, 101 S.Ct. 2860 (quoting Chimel, 395 U.S. at 763, 89 S.Ct. 2034). The Court therefore held that “when a policeman has made a lawful custodial arrest of the occupant of an automobile, he may, as a contemporaneous incident of that arrest, search the passenger compartment of that automobile” and any containers found within the passenger compartment. Id. ¶ 17 The “bright-line” rule established in Belton thus relieved police officers from a case-by-case justification as to whether the entire passenger compartment of a vehicle, as opposed to only a portion of the compartment, was within the “immediate control” of an arrestee who had been a recent occupant of the vehicle. But, although Belton provided clear guidance with respect to this spatial limitation on the incident to arrest exception in the context of a vehicle search, it did not specifically address a number of other questions. While noting that it was undertaking to provide some clarity as to the scope of a search incident to arrest when the defendant was a “recent occupant” of a vehicle, id. at 460, 101 S.Ct. 2860, the Court did not undertake to define recent occupancy, other than to note in the case before it that the search occurred immediately after the arrest and that Belton was a passenger in the car “just before he was arrested.” Id. at 462, 101 S.Ct. 2860. And, while stating generally that “a lawful custodial arrest creates a situation which justifies the contemporaneous search without a warrant of the person arrested and of the immediately surrounding area,” id. at 461, 101 S.Ct. 2860, Justice Stewart did not purport to set forth any rule as to where the arrestee must be located in relation to the vehicle at the time of arrest in order to justify a warrantless search of the passenger compartment. Again, this was likely because the defendant in Belton was in close proximity to the car at the time of arrest. See id. at 456, 101 S.Ct. 2860. ¶ 18 Justice Stewart made plain in Belton, however, that the Court was not retreating from Chimel, but rather simply applying its principles to the particular problem before it. See Belton, 453 U.S. at 460 n. 3, 101 S.Ct. 2860 (“[Belton] in no way alters the fundamental principles established in the Chimel case regarding the basic scope of searches incident to lawful custodial arrests.”). Thus, Belton carefully distinguished cases such as United States v. Chadwick, 433 U.S. 1, 97 S.Ct. 2476, 53 L.Ed.2d 538 (1977), in which the search of a footlocker occurred an hour after federal agents gained exclusive control of the item and after the defendants were securely in custody, and Arkansas v. Sanders, 442 U.S. 753, 99 S.Ct. 2586, 61 L.Ed.2d 235 (1979), which involved a suitcase in the trunk of a taxicab and thus not within the defendant’s “immediate control.” See Belton, 453 U.S. at 461-62, 101 S.Ct. 2860. In each case, Justice Stewart emphasized, there was no “arguably valid search incident to a lawful custodial arrest.” Id. at 462, 101 S.Ct. 2860. D. ¶ 19 Because neither Belton nor any subsequent Supreme Court case has defined just when a defendant is a sufficiently “recent occupant” of a vehicle so as to allow a search incident to arrest of the vehicle’s passenger compartment, state and federal courts have struggled to find a workable definition of the term. One line of cases has focused on the Court’s statement in Belton that its holding was limited to the “particular and problematic context” before it. Belton, 453 U.S. at 460 n. 3, 101 S.Ct. 2860. Noting that the police officer in Belton had ordered the driver of the vehicle to stop before the arrest occurred, see id. at 455, 101 S.Ct. 2860, a *164 number of courts have focused on whether the police had initiated contact with the arrestee while he was still in the vehicle. ¶ 20 United States v. Hudgins, 52 F.3d 115 (6th Cir.1995), exemplifies this approach. Hudgins held that as long as the police officer “initiates contact” with a defendant “by actually confronting the defendant or by signaling confrontation with the defendant, while the defendant is still in the automobile,” and the defendant is subsequently arrested, “a search of the automobile’s passenger compartment falls within the scope of Belton and will be upheld as reasonable.” Id. at 119. Conversely, if the defendant voluntarily left the automobile and began walking away before the police officer initiated contact, “the case does not fit within Belton’s bright-line rule, and a case-by-case analysis of the reasonableness of the search under Chimel becomes necessary.” Id. ¶ 21 Under this approach, which views a defendant as a “recent occupant” of a vehicle for purposes of the Belton rule only when the police initiated contact when the arrestee was still in the vehicle, state and federal courts have found Belton inapplicable when the defendant left the vehicle before such contact, even when the arrest occurred in close proximity to the car. See, e.g., United States v. Strahan, 984 F.2d 155 (6th Cir.1993); United States v. Fafowora, 865 F.2d 360 (D.C.Cir. 1989); Thomas v. State, 761 So.2d 1010 (Fla. 2000); People v. Femengel, 216 Mich.App. 420, 549 N.W.2d 361 (1996); Commonwealth v. Santiago, 410 Mass. 737, 575 N.E.2d 350 (1991). Division Two of our court of appeals recently adopted this approach in State v. Gant, 202 Ariz. 240, 244-45 ¶ 11, 43 P.3d 188, 192-93 (App.2002), cert. granted, — U.S. -, 123 S.Ct. 1784, 155 L.Ed.2d 664 (2003). ¶ 22 In Gant, the defendant had arrived in his car at a residence at which police already were present. Recognizing the defendant as someone wanted on an outstanding warrant and whose license was suspended, an officer arrested him after he exited his vehicle. The officer then searched the vehicle, discovering a weapon and cocaine inside the car. ¶ 23 The superior court held the search lawful as conducted incident to Gant’s arrest, but the court of appeals reversed. The court of appeals started from the proposition that the rule announced in Belton was limited to the “particular factual situation in which it arose.” Id. at 244 ¶ 11, 43 P.3d at 192. Because the police officer in Belton had initiated contact with the defendant while the defendant was still in the automobile, Gant concluded that Belton applied only under those circumstances — when “the officer initiates contact with the defendant, either by actually confronting the defendant or by signaling confrontation ... while the defendant is still in the automobile.” Id. (quoting Hudgins, 52 F.3d at 119 (alteration in original)). ¶ 24 In Gant, the police officer had not initiated contact while the defendant was still in the vehicle; the defendant drove the vehicle to the home where the officer already was present, and the confrontation first occurred after the defendant left the car. Id. at 242-43 ¶ 3, 43 P.3d at 190-91. Applying its interpretation of Belton, the court of appeals invalidated the search. Id. at 246 ¶ 15, 43 P.3d at 194. ¶ 25 In dictum, Gant “emphasize[d] that, when police attempt to initiate contact by either confronting or signaling confrontation, a vehicle’s occupant cannot avoid Belton’s application and create a haven for contraband by simply exiting the vehicle when officers are seen or approach.” Id. at 244-45 ¶ 11, 43 P.3d at 192-93. Under such circumstances, Gant suggested, the search is incident to an arrest when the suspect is “subsequently arrested.” Id. at 245 ¶ 11, 43 P.3d at 193. ¶ 26 Adopting the Gant dictum, the court of appeals in this case held that the search of the vehicle was incident to Dean’s arrest, notwithstanding that the arrest occurred long after he left the vehicle and when he was inside the house, because he fled the vehicle when the police approached after initiating contact: *165 Dean, 203 Ariz. at 412 ¶ 12, 55 P.3d at 106. The court of appeals reasoned that had Dean not fled the car at the approach of the police, he would have been arrested in or near the caí’, and any subsequent search of the vehicle would then have been plainly incident to the arrest. Id. at 412 ¶¶ 11-12, 55 P.3d at 106. Because both the time between Dean’s exit of the vehicle and arrest, and his distance from the vehicle at the time of the arrest resulted from Dean’s “attempt to evade” police, the court of appeals held that he could not “evade” the warrantless search of the Jeep. Id. at 412 ¶ 12, 55 P.3d at 106. *164 Dean cannot evade the search of the Jeep and the discovery of contraband in his vehicle by parking the Jeep and running into a house as soon as he is confronted by a police officer. The search, therefore, was incident to his arrest.... *165 E. ¶ 27 The analytic approach taken by the court of appeals in this case and in Gant, under which the applicability of the Belton rule turns entirely on whether the police initiated contact with the arrestee while he was still in the vehicle, is not supported by the rationale of either Belton or Chimel. The search incident to arrest exception explicated in Belton and Chimel was designed to protect officer safety and avoid the destruction of evidence. A suspect arrested next to a vehicle presents the same threat to officer safety and the same potential for destruction of evidence whether or not he was alerted prior to arrest of the police’s interest in him. Yet, under the approach Gant adopted from Hudgins and its progeny, the police would be able to search the entire passenger compartment of the automobile only if they initiate contact with a passenger while the suspect is in the vehicle; in all other cases, the Chimel “immediate control” test would apply. It makes no sense to have two different rules applicable to arrests occurring in what is for all relevant intents and purposes the same situation. ¶ 28 Moreover, the singular focus on whether the police initiated contact before the defendant departed the vehicle runs counter to one of the purposes underlying Belton. The Supreme Court sought in that case to create a workable definition of the area within the “immediate control” of a recent occupant of a vehicle, both to provide a “familiar standard” to police officers and to avoid case-by-case litigation as to whether the entire passenger compartment, or only a portion thereof, was within the scope of a search incident to arrest. See Belton, 453 U.S. at 458-60, 101 S.Ct. 2860. But, by defining “recent occupancy” entirely in terms of whether the defendant was formally made aware of the police presence before leaving the vehicle, Gant and the opinion below would return the courts to the very sort of inquiry that Belton abjured in every case in which contact was not so initiated before the arrest. 3 ¶ 29 For these reasons, a number of courts have found initiation of contact by the police irrelevant to the determination whether an arrestee was a “recent occupant” of a vehicle under Belton. See, e.g., United States v. Thornton, 325 F.3d 189, 194-95 (4th Cir.2003); United States v. Sholola, 124 F.3d 803, 817 (7th Cir.1997); Glasco, 513 S.E.2d at 141 — 42; see also Gant, 202 Ariz. at 244 n. 3, 43 P.3d at 192 n. 3 (recognizing a split in authority and collecting eases rejecting the “initiating contact” construction of Belton). We agree with the general analytical approach taken in these decisions. Belton made clear that its holding “in no way alters the fundamental principles established in the Chimel case.” Belton, 453 U.S. at 460 n. 3, 101 S.Ct. 2860. Thus, the appropriate inquiry focuses on the critical factors of when and where the custodial arrest took place. “[A] search can be incident to an arrest only if it is substantially contemporaneous with the arrest and is confined to the immediate vicinity of the arrest.” Stoner v. California, 376 U.S. 483, 486, 84 S.Ct. 889, 11 L.Ed.2d 856 (1964). *166 ¶30 The correct rule, we believe, was stated by the Supreme Court of Virginia in Glaseo. In upholding a vehicle search as incident to a lawful arrest, the Virginia court stated that a defendant is “a recent occupant of a vehicle within the limits of the Belton rule” when he is arrested “in close proximity to the vehicle immediately after the [defendant] exits the automobile.” Glaseo, 513 S.E.2d at 142. While concepts such as “close proximity” and “immediately after” are of course subject to factual analysis, they directly correspond to the rationales behind the search incident to arrest exception, which is designed to protect officer safety and preserve evidence. At the same time, the Glaseo test is faithful to the general notions that the Fourth Amendment disfavors warrant-less searches and that any exceptions to that general rule are narrowly limited in light of their underlying justifications. ¶ 31 In determining whether a search was “substantially contemporaneous” with an arrest of a recent occupant of a vehicle, the courts have reached a wide variety of results under a broad array of factual circumstances. Compare, e.g., United States v. Arango, 879 F.2d 1501, 1504-07 (7th Cir.1989) (upholding a search of a defendant’s car as incident to an arrest made of the defendant after he walked nine blocks away from the vehicle), with United States v. Vasey, 834 F.2d 782, 787-88 (9th Cir.1987) (invalidating a search of a vehicle occurring thirty to forty-five minutes after the defendant’s arrest); see also United States v. Strahan, 984 F.2d 155, 159 (6th Cir.1993) (holding that an arrestee was not a “recent occupant” of a vehicle when arrested thirty feet from the vehicle); People v. Savedra, 907 P.2d 596, 599 (Colo. 1995) (holding an arrestee to be a “recent occupant,” despite a five-minute lapse in time between the arrestee exiting the vehicle and the subsequent search); Gauldin v. State, 683 S.W.2d 411, 414 (Tex.Crim.App.1984) (holding that defendant who had parked truck and was sitting inside a restaurant when confronted by police was not a “recent occupant” of a truck); State v. Vanderhorst, 419 So.2d 762, 764 (Fla.Dist.Ct.App.1982) (finding defendant no longer a recent occupant where he had left his car, walked three miles home, requested a ride back to the car, and was attempting to rescue the car when police arrived). But we have been able to discover no case, and the State has cited none, in which a search of the passenger compartment of a vehicle was upheld under Belton when the driver was arrested as long after he left the vehicle and as far from vehicle as was the defendant here. ¶ 32 Under any reasoned analysis, Dean simply was not a “recent occupant” of the Jeep for Belton purposes when he was arrested. He had not occupied the vehicle for some two and one-half hours, and his arrest occurred not in close proximity to the car, but instead inside the house. Under the circumstances of this case, neither of the justifications for a warrantless search of the vehicle — protection of the arresting officers and preservation of evidence — is present. ¶ 33 While the bright-line rule announced in Belton relieves the police of demonstrating that a particular portion of the passenger compartment is within the “immediate control” of an arrestee, it does not purport to dispense with all such analysis as to whether the police may search the vehicle at all. Here, given the physical distance of the arrest ee from the vehicle at the time of arrest and the long lapse of time between the arrest and Dean’s exit from the vehicle, the search simply cannot be characterized as “incident” to the arrest and excepted from the general requirement that a warrant be obtained. ¶ 34 We therefore hold that when, as here, the arrest occurs long after the defendant had left the vehicle and far from the vehicle, the warrant requirement of the Fourth Amendment applies. Because, as Chimel and Belton teach, the Constitution requires a warrant except under those exigencies that allow otherwise, the issue is not whether the defendant has “evaded” a search by departing the vehicle, but rather whether the totality of the facts still presents the kind of situation that justifies dispensing with the warrant requirement. 4 In this case, the facts do not support such a result. *167 hi. ¶ 35 For the foregoing reasons, we conclude that the warrantless search of Dean’s Jeep cannot be justified as a search incident to arrest. Accordingly, we vacate the opinion of the court of appeals and reinstate the order of the superior court suppressing the evidence seized in this search. This case is remanded to the superior court for further proceedings consistent with this opinion. CONCURRING: CHARLES E. JONES, Chief Justice, RUTH V. McGREGOR, Vice Chief Justice, REBECCA WHITE BERCH, and Michael D. Ryan, Justices. Footnotes [1] While Dean argues that the search in this case violates both the Fourth Amendment and Article 2, Section 8 of the Arizona Constitution, he presents no separate arguments based on the state constitutional provision. We therefore address his claim only under the United States Constitution. See State v. Nunez, 167 Ariz. 272, 274 n. 2, 806 P.2d 861, 863 n. 2 (1991). [2] Nor does this case involve the so-called "automobile” exception to the warrant requirement of the Fourth Amendment. Under that exception, searches of vehicles may be allowed absent a warrant if the police have "probable cause” to do so. See Chambers v. Maroney, 399 U.S. 42, 90 S.Ct. 1975, 26 L.Ed.2d 419 (1970). The State candidly conceded at oral argument that the record in this case does not establish probable cause for the search of the Jeep, and thus did not argue in this court that the "automobile" exception applies. [3] Michigan v. Long, 463 U.S. 1032, 103 S.Ct. 3469, 77 L.Ed.2d 1201 (1983), suggests in dictum that the Supreme Court did not mean to limit Belton to cases in which the police initiated contact while the defendant was in the vehicle. That case involved a defendant who met police officers at the scene of an accident, where his vehicle had swerved off the road into a ditch. In that case, in upholding a search of the vehicle under the principles set forth in Terry v. Ohio, 392 U.S. 1, 88 S.Ct. 1868, 20 L.Ed.2d 889 (1968), the Court also specifically noted that if the officers had arrested the defendant for various traffic violations, they could have searched the passenger compartment of the car under Belton. Long, 463 U.S. at 1035 n. 1, 103 S.Ct. 3469. [4] Whether the defendant "fled” the vehicle is no more relevant to the Belton analysis than a defendant's flight from a home would be relevant to the determination of whether a search of that *167 home after the defendant was arrested miles away was incident to that arrest. Indeed, by its very nature, flight from the vehicle tends to minimize the dual concerns that underlie the search incident to arrest exception — police safety and protection of evidence. Flight may, of course, be relevant to establish other exceptions to the Fourth Amendment warrant requirement, such as whether the defendant has abandoned the vehicle, whether an administrative inventory of the vehicle is necessary, or whether sufficient probable cause exists to justify a warrantless search under the automobile exception.