¶ 12. dissenting. As Professor Wayne LaFave has accurately observed, “if the ultimate issue is perceived as whether the suspect “would feel free to walk away,’ then virtually all police-citizens encounters must in fact be deemed to involve a Fourth Amendment seizure.” 4 W. LaFave, Search and Seizure § 9.3(a), at 99 (3d ed. 1996) (quoting State v. Evans, 517 P.2d 1225, 1229 (Or. Ct. App. 1974)). In continuing down the line of labeling less and less intrusive interactions as seizures, the majority has literally applied the wording of the Fourth Amendment test, but not its content as it has evolved in decisions from the United States Supreme Court and other courts. As a result, it has ignored *517Professor LaFave’s characterization of its policy and removed from law enforcement desirable options to protect vulnerable citizens. For this reason, I dissent.
¶ 13. I think that the correct line between appropriate police-citizen interaction and a police seizure of a citizen can be found in two opinions of the United States Supreme Court, Florida v. Bostick, 501 U.S. 429 (1991), and United States v. Drayton, 536 U.S. 194 (2002), and that line would require us to affirm the decision of the trial court in this case. Both Bostick and Drayton involve police encounters with bus passengers under programs in which law enforcement officials board busses during scheduled depot stops and interrogate passengers and request to search their luggage for drugs. In Bostick, a passenger on whom the officers found drugs challenged the practice, and the Florida Supreme Court found the practice per se unconstitutional because the passengers were not realistically free to leave the bus to avoid the questioning and searches. 501 U.S. at 433. In an argument comparable to that made here “Bostick insisted] that this ease is different because it took place in the cramped confines of a bus. A police encounter is much more intimidating in this setting,... because police tower over a seated passenger and there is little room to move around.” Id. at 435. The Supreme Court disagreed noting that the passengers were not free to leave because they were on a bus that was departing from the depot, and not because of law enforcement coercion. Id. at 436-37. Thus, the test was whether the passengers were free to decline the request to search “or otherwise terminate the encounter.” Id. The Court held the test required the Florida courts to consider all circumstances, including the facts that the officers did not use their firearms and they specifically advised each passenger that consent to search was voluntary, and remanded for that purpose. Id. at 437-38.
¶ 14. Drayton involved almost the same facts except that the officer did not advise the passenger that consent to search was voluntary. 536 U.S. at 197. Also in Drayton, three officers were stationed on the bus — one at the rear, one at the entrance and the last interrogating the passengers but not blocking the aisle. The officer asking questions was twelve to eighteen inches from the passenger. The Supreme Court upheld the district court’s finding that the passengers were not seized:
The officers gave the passengers no reason to believe that they were required to answer the officers’ questions. When Officer Lang approached respondents, he did not brandish a weapon or make any intimidating movements. He left the aisle free so that respondents could exit. He spoke to passengers one by one and in a polite, quiet voice. Nothing he said would suggest to a reasonable person that he or she was barred from leaving the bus or otherwise terminating the encounter.
... There was no application of force, no intimidating movement, no overwhelming show of force, no brandishing of weapons, no blocking of exits, no threat, no command, not even an authoritative tone of voice.
Id. at 203-04.
¶ 15. We must look at the seizure standard based on the assumption that the citizen who interacts with the police is innocent of criminal behavior. See Bostick, 501 U.S. at 438. We cannot base our decision on the natural inclination of citizens to cooperate with police requests. See People v. Melton, 910 P.2d 672, 677 (Colo. 1996).
*518¶ 16. Of course, the criticism of Dray-ton and Bostick is that the test espoused does not match the actions of the police upheld in the decisions. See generally J. Nadler, No Need to Shout: Bus Sweeps and the Psychology of Coercion, 2002 Sup. Ct. Rev. 153. Whatever the general language used, the decisions of the United States Supreme Court continue to move “away from labeling such minor restrictions on individual movement as seizures.” State v. Burgess, 163 Vt. 259, 263, 657 A.2d 202, 204 (1995) (Dooley, J., dissenting). Drayton and Bostick give little weight to the kind of considerations the majority has focused on here, particularly the restrictions on movement, and instead focus on the factors itemized in United States v. Mendenhall, 446 U.S. 544, 554-55 (1980):
Examples of circumstances that might indicate a seizure, even where the person did not attempt to leave, would be the threatening presence of several officers, the display of a weapon by an officer, some physical touching of the person of the citizen, or the use of language or tone of voice indicating that compliance with the officer’s request might be compelled____ In the absence of some such evidence, otherwise inoffensive contact between a member of the public and the police cannot, as a matter of law, amount to a seizure of that person.
(Citations omitted.)
¶ 17. The majority ignores the Mendenhall factox'S, apparently seeing this case as the logical extension of Burgess, in which this Could; held that one factor alone — an officer’s use of official, flashing blue lights to approach a parked automobile — created a seizure because the lights tend “to inhibit a suspect’s departure from the scene.” 163 Vt. at 261, 657 A.2d at 203. Burgess’s depai-ture from a weighing of all the factor's is understandable, as many cases have held, because state law typically requires a motorist to remain stopped when the flashing emergency lights are illuminated. See Hrezo v. State, 780 So. 2d 194, 195 (Fla. Dist. Ct. App. 2001); State v. Mireles, 991 P.2d 878, 880 (Idaho Ct. App. 1999); State v. Morris, 72 P.3d 570, 577 (Kan. 2003). There is no equivalent reason to ignore all the factors, particularly those outlined in Mendenhall, in this case and to give controlling weight to only the two factors considered by the majority.
¶ 18. The majority also di'aws from the evidence its version of the facts and ignores the findings of the trial judge. The court found:
Nothing was blocking her way initially when the officer was standing by the car and she could have pulled behind him in reverse and then gone forward directly out of the lot and that she was not trapped in there in any fashion by where the officer’s car was parked or by where he was standing. So I don’t conclude that a reasonable person would have felt that they could not get out of that lot if they chose to do so and if they had said “I’d rather not answer any questions.”
(Emphasis added.) We are bound by the trial court’s finding of the underlying facts. See State v. Lawrence, 2003 VT 68, ¶¶ 8-9, 175 Vt. 600, 834 A.2d 10 (mem.). By l'elying on some of the evidence, leather than the findings, the majoi’ity overstates the degree to which the freedom of movement of the vehicle was impeded. Although we review the trial court’s conclusion de novo, we must do so consistent with the findings.
*519¶ 19. The proper analysis, based on the findings made by the trial court, is demonstrated by two leading cases from other jurisdictions that analyze in some detail the circumstances where the officer’s vehicle is parked to partially block the defendant’s vehicle, what the trial court found occurred here. The first is United States v. Kim, 25 F.3d 1426 (9th Cir. 1994), in which a DEA agent positioned his unmarked car to partially block a drug suspect’s car containing defendant while it was parked on the street in front of a store. The appeals court upheld the district court ruling that no seizure had occurred. After reviewing the general law, the court addressed defendant’s argument that the partial blocking of his car created a seizure:
To be sure, where officers detain an already stationary suspect by hindering his future as opposed to ongoing progress, that they did not stop the suspect as the term is commonly understood does not foreclose inquiry into whether their conduct constitutes an investigatory stop. However, that [the officer] partially blocked Kim’s egress with his automobile informs but does not alter our conclusion that Kim was not stopped in the constitutional sense before his surrender of the vial setting the foundation for the subsequent search____ The phrasing of [the officer’s] request for permission to question Kim left open the possibility of a refusal and the positions in which DEA agents were posted did not entirely bar Kim’s egress.
Id. at 1431 (citations omitted); see United States v. Bates, No. 01-30199, 2002 WL 31119844, at *1 (9th Cir. Sept. 25, 2002); United States v. Summers, 268 F.3d 683, 687 (9th Cir. 2001); McCormick v. City of Lawrence, 253 F. Supp. 2d 1172, 1188-89 (D. Kan. 2003); United States v. Baldwin, No. Cr.3:97CR188(AHN), 1998 WL 563851, at *3 (D. Conn. June 29, 1994).
¶ 20. The second, People v. Cascio, 932 P.2d 1381 (Colo. 1997), is even more relevant because it involves facts very similar to those in this case. The officers in Cascio came across a van parked off an unpaved national forest road after dark. They parked then- vehicle to .partially block the van and illuminated it with a spotlight. After questioning the occupant, defendant, the officers found drugs. Id. at 1383. The Colorado Supreme Court discussed the general law and noted that where a vehicle is totally blocked the courts have found a seizure. Id. at 1387. It held that the law was different when the vehicle was partially blocked so “that the Cascios would have been able to leave by maneuvering their van in a manner akin to parallel parking.”2 Id. In that case, defendant was “not physically restrained from departing,” and the partial *520blocking should be considered only one factor to be weighed in the totality of the circumstances. Id. The court found no seizure with the following analysis:
While this factor [the partial, blocking], standing alone, is not controlling, the totality of the circumstances surrounding the encounter between the Cascios and the sheriff’s deputies does not support a finding that the encounter was an investigatory stop rather than a consensual encounter. Although there were two deputies present, they did not act in a threatening manner. In particular, the deputies did not display their weapons, physically touch the Cascios, surround the Cascios, or use an intimidating tone of voice. In fact, the colloquy between the parties was rather mild-mannered. Deputy Rosenbaum’s approach was non-threatening. “His greeting to the Cascios — Hi guys. How are you doing?” — was casual and friendly. Deputy Rosenbaum testified that he pulled over with the intention of “contacting” the Cascios. Deputy Getskow testified that he and Deputy Rosenbaum “were told to investigate vehicles that were off the road, and that was basically for welfare checks, and so forth.” Indeed, as we noted in People v. Chaves, 855 P.2d 852 (Colo. 1993), police officers have various roles and multiple tasks in addition to those related to criminal offenses, including some that are civil in nature.
Id. at 1387-88 (citations omitted).
¶ 21. In addition we note that in the numerous cases where officers have used lights, other than official flashing emergency lights, to illuminate the interior of a parked car, the courts have not given controlling significance to that factor. Adams v. State, 758. S.W.2d 709, 712 (Ark. Ct. App. 1988) (spotlight); People v. Paynter, 955 P.2d 68, 73 (Colo. 1998) (spotlight); State v. O’Neill, 62 P.3d 489, 497 (Wash. 2003) (spotlight and flashlight). Indeed, Cascio, like this case, involved both partial blocking and a spotlight to illuminate the interior of the parked vehicle.
¶ 22. The other relevant factors in this case are all inconsistent with finding a seizure. There was only one officer. He did nothing threatening. He did not bring his weapon out. He did not touch anyone. There was no evidence that he used an intimidating or threatening tone of voice. He simply asked the occupants of the vehicle what they were doing. Based on these factors primarily, I would hold that no seizure occurred here.
¶ 23.1 agree with the majority that our precedents require that a seizure for community caretaking purposes be based on specific and articulable facts supporting the seizure and would agree normally that such a justification for the seizure before us cannot be upheld. But I also agree that the degree of intrusion is minimal compared to the kind of risk that motivated the officer to inquire about the circumstances of a couple in a vehicle in a very remote place at 2 A.M. in the morning.
¶ 24. As the majority continues to expand what conduct will be considered a seizure when a police officer approaches a stopped vehicle after dark, we are left with only two options: the officer will approach the vehicle giving no visible sign that the person doing so is a police officer rather than a robber or other person engaged in criminal conduct; or the officer will ignore the vehicle. The first option greatly places at risk the safety of the officer and the occupants of the vehicle. The latter eliminates the ability of the officer to protect public *521safety under circumstances that should warrant some inquiry. See State v. Walters, 934 P.2d 282, 288 (N.M. Ct. App. 1997) (lcWe are loathe to discourage community caretaker stops or to make them hazardous for motorists or the police.”); State v. Mireles, 991 P.2d at 881-82 (same). I cannot agree that these should be the only options and therefore dissent.
¶25. I am authorized to state that Justice Reiber joins in this dissent.
The majority’s attempt to distinguish the facts in Cascio is unconvincing. The van in that ease “was parked a few feet away from ... boulders,” 932 P.2d 1381, 1383 (Colo. 1997), when the police car pulled ten to twenty feet in back of it and shined its spotlight on the occupants. In order to leave, the van had to back around the police car “in a manner akin to parallel parking,” and then go forward. Id. at 1387. Based on the facts found by the trial court here, the driving necessary to exit from the trailhead parking lot was similar, if not identical. Apparently, the majority finds that headlights shined into a car are more intrusive than a spotlight shined on the occupants. The majority calls this a subtle distinction. I would call it a distinction with no relevant difference.