¶1 Inside defendant Kevin McKnight's truck, police officers discovered a pipe containing what later proved to be methamphetamine residue. That discovery culminated in McKnight's conviction for certain drug offenses. On appeal, he challenged the constitutionality of the search that revealed the pipe.

¶2 A division of the court of appeals reversed McKnight's convictions. People v. McKnight , 2017 COA 93, ¶ 25, --- P.3d ----. Each member of the division wrote separately, wrestling with what effect, if any, legalized marijuana in Colorado should have on the constitutionality of the search of McKnight's truck. To which you might say, but this is a methamphetamine case. So, why are we talking about marijuana? Well, the drug-detection dog used to find the pipe, Kilo, was trained to alert on multiple drugs, including marijuana. Even a hint of marijuana can trigger the same response from Kilo as any quantity of methamphetamine.

¶3 And that's where things get tricky. After all, the possession of an ounce or less of marijuana by someone twenty-one or older is legal in Colorado, following the passage of Amendment 64, Colo. Const. art. XVIII, § 16 (3), even though such possession remains illegal under federal law. Thus, no matter how reliable his nose, Kilo can now render a kind of false positive for marijuana. He has been trained to alert to marijuana based on the notion that marijuana is always contraband, when that is no longer true under state law. And historically, whether a drug-detection dog might alert on noncontraband drives whether the dog's sniff constitutes a search implicating constitutional protections. The dog's sniff arguably intrudes on a person's reasonable expectation of privacy in lawful activity. If so, that intrusion must be justified by some degree of particularized suspicion of criminal activity.

¶4 Does this mean that Amendment 64 gave McKnight a reasonable expectation of privacy under either the federal or state constitution that was violated by Kilo's keen sense of smell? McKnight says yes. Among other things, he claims that the Colorado Constitution prohibited Kilo's intrusion without at least reasonable suspicion that McKnight had committed or was committing a crime. Because there was none, McKnight asserts that the trial court should have suppressed the pipe. Two members of the division agreed. McKnight , ¶ 3.

¶5 And even if the "sniff [was] up to snuff" (to use Justice Kagan's popular shorthand from a slightly different context in Florida v. Harris , 568 U.S. 237, 248, 133 S.Ct. 1050, 185 L.Ed.2d 61 (2013) ), McKnight further argued that there was no probable cause for the post-sniff hand search of his truck that revealed the pipe. Again, because of Amendment 64, two members of the division agreed. McKnight , ¶ 4.

¶6 The People counter that, despite Amendment 64, marijuana remains contraband in many circumstances at the state level, and illegal under all circumstances federally, and thus Kilo's sniff was not a search requiring so much as reasonable suspicion. The People further contend that an alert from a dog trained to detect marijuana, in *400addition to other substances, still provides probable cause justifying a search.

¶7 We hold that a sniff from a drug-detection dog that is trained to alert to marijuana constitutes a search under article II, section 7 of the Colorado Constitution because that sniff can detect lawful activity, namely the legal possession of up to one ounce of marijuana by adults twenty-one and older. We further hold that, in Colorado, law enforcement officers must have probable cause to believe that an item or area contains a drug in violation of state law before deploying a drug-detection dog that alerts to marijuana for an exploratory sniff. Because there was no such probable cause justifying Kilo's search of McKnight's truck, the trial court erred in denying McKnight's motion to suppress.

¶8 Accordingly, we affirm the judgment of the court of appeals.

I. Facts and Procedural History

¶9 Evidence presented at a suppression hearing established the following facts.

¶10 While on patrol in an unmarked police vehicle one night in February 2015, Craig Police Officer Bryan Gonzales observed a parked pickup truck facing the wrong way in a one-way alley near an apartment complex. A man stood outside of the passenger-side door of the truck. Although Officer Gonzales saw no behavior consistent with an exchange or transaction, he followed the truck as it traveled a few blocks. The truck then parked in front of a residence where police had found drugs almost two months earlier, and it remained parked there for approximately fifteen minutes. During that time, no one exited the truck or the residence. When the truck started moving again, Gonzales trailed along.

¶11 When the driver of the truck failed to signal a turn, Gonzales pulled the truck over. He discovered that McKnight was the driver. And Gonzales recognized the passenger as someone who had used methamphetamine "at some point in the past," but he wasn't sure how recently.

¶12 During the traffic stop, Gonzales asked Sergeant Courtland Folks of the Moffat County Sheriff's Office to respond with his drug-detection dog, Kilo, who had been trained to detect the odors of marijuana, methamphetamine, cocaine, heroin, and ecstasy. If Kilo detects the scent of any one of those substances, he should alert. And he exhibits the same alert for all five drugs.

¶13 Within five minutes of the stop, Kilo was on the job. As Folks walked Kilo around McKnight's truck, Folks asked McKnight if he had "any narcotics on him." McKnight said no. Kilo quickly "alerted" on the driver's door beneath the driver's open window. This means that Kilo engaged in a rapid sniffing pattern and behaved in a manner suggesting the presence of one of the substances on which he had been trained. According to Folks, Kilo then "put his nose on the driver's door, back to the door handle, did a purge, which means he cleared his nose, took another deep breath and immediately started giving a trained indication, which was barking."

¶14 The officers then ordered McKnight and the passenger to exit the truck. After they complied, the officers patted them down and found nothing on them. The officers then searched the truck by hand. In a storage compartment under the rear seat, they found a pipe containing suspected methamphetamine residue.

¶15 Before trial, the defense moved to suppress the pipe, arguing that it was the fruit of an unconstitutional search. Specifically, McKnight argued that because marijuana is legal in Colorado and Kilo was trained to detect marijuana, Kilo's sniff was a search that required particularized suspicion of criminal activity before Kilo could be deployed, and there was no such suspicion on these facts. Moreover, he contended that Kilo's alert did not provide probable cause for a full-blown, human search of the truck. Through counsel, McKnight grounded his arguments in the search and seizure clauses of both the United States and Colorado Constitutions.

¶16 After a hearing, the trial court denied the motion to suppress. In relevant part, the trial court noted that the possession of marijuana remains illegal under many circumstances in Colorado and is categorically illegal *401under federal law. Without explicitly stating as much, the trial court seemed to reason that a sniff by a dog trained to detect marijuana does not constitute a search. And even if Kilo's sniff was a search, the trial court went on to conclude that there was reasonable suspicion of criminal activity supporting the search. With ample record support, the court also concluded that Kilo reliably sniffs out the drugs on which he's been trained. The trial court did not address whether there was probable cause for the officers' hand search of the truck.

¶17 A jury later convicted McKnight of possession of a controlled substance and possession of drug paraphernalia.

¶18 McKnight appealed. In three separate but partially overlapping opinions, the court of appeals unanimously agreed that the trial court erred in denying McKnight's motion to suppress the pipe and that the error was not harmless. Therefore, the division reversed McKnight's convictions.

• Judge Dailey, writing for himself and Judge Berger, agreed with McKnight that, under the Colorado Constitution after the enactment of Amendment 64, Kilo's sniff was a search requiring reasonable suspicion of criminal activity. McKnight , ¶ 18 ("Because a dog sniff of a vehicle could infringe upon a legitimate expectation of privacy solely under state law, that dog sniff should now be considered a 'search' for purposes of article II section 7 of the state constitution where the occupants are twenty-one years or older."). Judge Dailey further determined that the reasonable suspicion standard should govern, and he concluded that the totality of the circumstances did not give the police reasonable suspicion that McKnight had engaged in criminal activity. Id. at ¶¶ 19-20, 22-24.

• Judge Jones, writing for himself and Judge Berger, agreed with McKnight that Kilo's alert, even in combination with other circumstances, did not give the police probable cause to conduct a warrantless hand search of McKnight's truck. Id. at ¶¶ 49-54 (Jones, J., specially concurring). In doing so, Judge Jones sidestepped the issue of whether Kilo's sniff now constitutes a search under state law. Id. at ¶ 37.

• Judge Berger, writing for himself, explained how a person could have an enforceable expectation of privacy in the possession of marijuana under state law, even in the face of a federal prohibition. First, he noted that Colorado courts must give effect to Colorado voters' enactment of Amendment 64. Id. at ¶ 29 (Berger, J., specially concurring). Second, he observed that the People have not argued that Amendment 64 is preempted by federal law. Id. Finally, he concluded that when Amendment 64 legalized possession of up to one ounce of marijuana for personal use by persons twenty-one years of age or older, it also limited police authority to enforce the federal prohibition. Id. at ¶ 31.

¶19 The People filed a petition for a writ of certiorari, and we agreed to review the case.1

II. Analysis

¶20 After identifying the standard of review, we examine the parallel evolution of federal and state jurisprudence concerning police use of drug-detection dogs. We review long-standing law reasoning that a sniff is not a search because an individual has no legitimate expectation of privacy as to contraband. But after the passage of Amendment 64, possession of an ounce or less of marijuana by someone twenty-one or older is legal, and thus, marijuana is no longer always "contraband" under state law. Because Kilo's sniff could detect lawful activity, we conclude his sniff was a search under the Colorado Constitution. We then consider *402what level of suspicion would justify deploying such a dog to sniff a vehicle, ultimately concluding that, because the sniff is a search, there must be probable cause to believe a vehicle contains illegal narcotics under state law before deploying a drug-detection dog trained to alert to marijuana. We conclude that, under the totality of the circumstances, there was no probable cause justifying the use of Kilo to sniff McKnight's truck or the subsequent hand search of the truck, and, exclusion of the pipe is the appropriate remedy for this violation of the Colorado Constitution.

A. Standard of Review

¶21 When reviewing a suppression order, we defer to the trial court's factual findings if the record supports them, but we review de novo the court's legal conclusions. Grassi v. People , 2014 CO 12, ¶ 11, 320 P.3d 332, 335. So, we address the constitutionality of Kilo's sniff and the subsequent hand search of the truck de novo.

B. Federal Law Governing Searches and Drug-Detection Dogs

¶22 The Fourth Amendment to the United States Constitution states:

The right of the people to be secure in their persons, houses, papers, and effects, against unreasonable searches and seizures, shall not be violated, and no Warrants shall issue, but upon probable cause, supported by Oath or affirmation, and particularly describing the place to be searched, and the persons or things to be seized.

It "protects people, not places" from unreasonable governmental searches and seizures. Katz v. United States , 389 U.S. 347, 351, 88 S.Ct. 507, 19 L.Ed.2d 576 (1967) ; see also U.S. Const. amend. IV. A search in the constitutional sense occurs "when the government violates a subjective expectation of privacy that society recognizes as reasonable." Kyllo v. United States , 533 U.S. 27, 33, 121 S.Ct. 2038, 150 L.Ed.2d 94 (2001) (citing Katz , 389 U.S. at 361, 88 S.Ct. 507 (Harlan, J., concurring)).

¶23 "Warrantless searches are presumptively unreasonable," and thus unconstitutional, unless an exception to the warrant requirement exists. United States v. Karo , 468 U.S. 705, 717, 104 S.Ct. 3296, 82 L.Ed.2d 530 (1984). Courts measure the constitutionality of a search (either pursuant to a warrant or warrantless) at the time a search is performed. The Fourth Amendment does not favor post hoc justifications based on what was found during a search. See, e.g. , People v. Sotelo , 2014 CO 74, 336 P.3d 188 ("[C]ourts may not use the benefit of hindsight in evaluating application of the Fourth Amendment."); see also California v. Acevedo , 500 U.S. 565, 599, 111 S.Ct. 1982, 114 L.Ed.2d 619 (1991) (Stevens, J., dissenting) ("Neither evidence uncovered in the course of a search nor the scope of the search conducted can be used to provide post hoc justification for a search unsupported by probable cause at its inception."); United States v. Montoya de Hernandez , 473 U.S. 531, 559, 105 S.Ct. 3304, 87 L.Ed.2d 381 (1985) (Brennan, J., dissenting) ("[P]ost hoc rationalizations have no place in our Fourth Amendment jurisprudence, which demands that we 'prevent hindsight from coloring the evaluation of the reasonableness of a search or seizure.' " (quoting United States v. Martinez-Fuerta , 428 U.S. 543, 565, 96 S.Ct. 3074, 49 L.Ed.2d 1116 (1976) )).

¶24 A warrantless search may be constitutional if there was probable cause to believe the place or item to be searched contained contraband or evidence of a crime, and the circumstances met an exception to the warrant requirement. One such exception is the automobile exception. Beginning with Carroll v. United States , the Supreme Court has reasoned that automobiles warrant a lesser degree of privacy than that afforded to homes because of a vehicle's "ready" mobility, the existing, pervasive government regulation of cars, and the low probability that a car is used to store personal effects. See 267 U.S. 132, 153, 45 S.Ct. 280, 69 L.Ed. 543 (1925) (reasoning that there is "a necessary difference between a search of a store, dwelling house, or other structure ... [where] a proper official warrant readily may be obtained and a search of a ship, motor boat, wagon, or automobile ... where it is not practicable to secure a warrant, because the *403vehicle can be quickly moved"). Nearly a century later, the law remains the same: If an officer has lawfully stopped a vehicle, and has probable cause to believe the vehicle contains evidence of a crime, then the officer may conduct a search of the car without first obtaining a warrant. Collins v. Virginia , --- U.S. ----, 138 S. Ct. 1663, 1670, 201 L.Ed.2d 9 (2018).

¶25 As the Supreme Court continued to embrace and refine the automobile exception, drug-detection dogs scampered on to the scene. United States v. Place marked the Supreme Court's first foray into this area of the law. 462 U.S. 696, 103 S.Ct. 2637, 77 L.Ed.2d 110 (1983). In Place , two law enforcement officers grew suspicious that Place might be carrying narcotics after observing him at Miami International Airport. Id. at 698, 103 S.Ct. 2637. The officers remained suspicious after talking with the defendant, though they ultimately decided not to search his luggage. Id. The officers looked closely at the luggage tags on the defendant's checked bags and noticed discrepancies between the two addresses and the phone number the defendant gave the airline. Id. The officers notified law enforcement agents in New York of the defendant's impending arrival. Id. Two Drug Enforcement Administration agents contacted the defendant when he landed, informed the defendant that they believed he may be transporting illegal narcotics, and ultimately seized his luggage in order to subject it to a sniff from a drug-detection dog. Id. at 698-99, 103 S.Ct. 2637.

¶26 Though Place turned on whether the seizure of the defendant's luggage without probable cause violated the Fourth Amendment, the Court discussed the use of drug-detection dogs at airports to sniff luggage suspected of containing illegal narcotics. See id. at 707, 103 S.Ct. 2637. The Court reasoned that exposing "temporarily detain[ed] personal luggage" to a dog trained to detect narcotics wasn't a search under the Fourth Amendment. Id. at 697-98, 103 S.Ct. 2637. The sniff is "sui generis," stated the Court, because it is less intrusive than other search techniques and it can only reveal the "presence or absence of narcotics, a contraband item." Id. at 707, 103 S.Ct. 2637. The Court's rationale thus hinged on two notions: (1) A dog sniff is a minimally intrusive tactic (2) that can only reveal whether or not contraband is present in the luggage. Beyond that, the sniff reveals nothing. Despite the narrow circumstances in which the Court made these statements, they have formed the foundation for much federal (and state) jurisprudence on dog sniffs.

¶27 The Supreme Court relied on this reasoning in Illinois v. Caballes to conclude that "the use of a well-trained narcotics-detection dog-one that 'does not expose noncontraband items that otherwise would remain hidden from public view'-during a lawful traffic stop, generally does not implicate legitimate privacy interests." 543 U.S. 405, 409, 125 S.Ct. 834, 160 L.Ed.2d 842 (2005) (internal citation omitted) (quoting Place , 462 U.S. at 707, 103 S.Ct. 2637 ). In Caballes , an officer led a drug-detection dog around a vehicle that was lawfully stopped for speeding. Id. at 406, 125 S.Ct. 834. The dog alerted at the trunk, leading to a search that revealed marijuana. Id. Looking back to Place , the Court rejected the claim that the dog sniff was unconstitutional because there was no reasonable suspicion that the defendant possessed narcotics when he was initially stopped. Id. at 408-09, 125 S.Ct. 834. Notably, the Court didn't focus on how the officers in Place had already grown suspicious that the defendant was engaged in illegal narcotics activity before stopping and contacting the defendant. Compare id. , with Place , 462 U.S. at 698-700, 103 S.Ct. 2637. Instead, the Court reasoned that reasonable suspicion wasn't necessary to justify the dog sniff because it could only detect contraband, and "any interest in possessing contraband cannot be deemed 'legitimate,' ... thus, governmental conduct that only reveals the possession of contraband 'compromises no legitimate privacy interest.' " Id. at 408, 125 S.Ct. 834 (quoting United States v. Jacobsen , 466 U.S. 109, 123, 104 S.Ct. 1652, 80 L.Ed.2d 85 (1984) ).2

*404C. Colorado Law: Pre-Amendment 64

¶28 Our search and seizure jurisprudence took a different path. In decades past, our opinions demonstrated a willingness to interpret the state constitution to afford broader protections than its federal counterpart. Yet, in recent years, we have moved away from this interpretive independence.

¶29 Article II, section 7 of the Colorado Constitution states:

The people shall be secure in their persons, papers, homes and effects, from unreasonable searches and seizures; and no warrant to search any place or seize any person or things shall issue without describing the place to be searched, or the person or thing to be seized, as near as may be, nor without probable cause, supported by oath or affirmation reduced to writing.

¶30 In Charnes v. DiGiacomo , this court first charted its own course in interpreting article II, section 7 to afford broader protections than the Fourth Amendment. 200 Colo. 94, 612 P.2d 1117 (1980). Charnes concerned a taxpayer's challenge to a subpoena issued to his bank by the Colorado Department of Revenue seeking the taxpayer's financial records. Id. at 1119. Applying United States v. Miller , this court concluded that the Fourth Amendment did not shield the taxpayer's financial records from government seizure because the taxpayer could not claim any expectation of privacy in the information he shared with his bank through financial transactions. Id. at 1120 (citing 425 U.S. 435, 96 S.Ct. 1619, 48 L.Ed.2d 71 (1976) ). But, the analysis did not end there. We then looked to article II, section 7 and concluded that the analytical framework announced in Katz -whether the defendant had a subjective expectation of privacy in his bank records that society was prepared to recognize as reasonable-was also the framework to be applied under the Colorado Constitution. See id. at 1120-21. Noting that Miller "does not determine the scope of protection provided to individuals in Colorado by the constitution of this state," we concluded that, under the Colorado Constitution, bank depositors have a reasonable expectation of privacy in records concerning their financial transactions held at banks in Colorado. Id. Charnes was only the first in a series of cases holding that article II, section 7 provides greater privacy protections than the Fourth Amendment. See, e.g. , People v. Sporleder , 666 P.2d 135, 143-44 (Colo. 1983) (rejecting the U.S. Supreme Court's reasoning in Smith v. Maryland , 442 U.S. 735, 99 S.Ct. 2577, 61 L.Ed.2d 220 (1979), to hold that the installation and use of a pen register constitutes a search under state law); People v. Corr , 682 P.2d 20, 27-28 (Colo. 1984) (applying the reasoning of Sporleder to conclude that a warrant is required to obtain telephone toll records); People v. Oates , 698 P.2d 811, 815-16 (Colo. 1985) (rejecting the reasoning of Karo , 468 U.S. 705, 104 S.Ct. 3296, and holding that the installation and presence of a tracking device in purchased commercial goods constitutes a search under the Colorado Constitution).

¶31 In People v. Unruh , we considered the constitutional permissibility of a dog sniff of a locked safe. 713 P.2d 370, 378-79 (Colo. 1986), abrogated by People v. Esparza , 2012 CO 22, 272 P.3d 367. We concluded that, despite the less intrusive nature of a dog sniff, it was a search because the defendant had a reasonable expectation of privacy in the locked safe. Id. In coming to this conclusion, we distinguished Place , noting that the sniff of a person's luggage at an airport was not deemed a search because the person has "no reasonable expectation of privacy from an investigative technique as non-intrusive as a dog sniff" at the airport. Id. at 378. We then had to decide whether a warrant was necessary to deploy the drug-detection dog, or whether "something less than probable cause" could justify the sniff. Id. We looked to Terry v. Ohio , 392 U.S. 1, 30-31, 88 S.Ct. 1868, 20 L.Ed.2d 889 (1968), where the U.S. Supreme Court held that an officer may conduct a stop if he has reasonable suspicion criminal activity is occurring. We analogized *405an exploratory dog sniff to the brief investigatory stops at issue in Terry , and we concluded that the proper "balance between governmental and individual interests in this case" necessitated that there be reasonable suspicion to justify the exploratory sniff. Unruh , 713 P.2d at 379. Our reasoning rested primarily on two principles: (1) the defendant had a reasonable expectation of privacy in the item searched (his locked safe), and (2) a dog sniff, though a search, was a less intrusive search than a traditional search by hand. Id. at 377-79.

¶32 In People v. Haley , we applied the reasoning of Unruh to conclude that a dog sniff resulting from an impermissibly prolonged traffic stop violated the protections of article II, section 7.3 41 P.3d 666, 673-74, 676-77 (Colo. 2001), abrogated by Esparza , 2012 CO 22, 272 P.3d 367. This was so because the defendants had "a privacy interest in their persons and vehicle being free from unreasonable governmental intrusion, and the drug investigation following the traffic stop in this case required reasonable suspicion for the dog sniff search to proceed." Id. at 674. We acknowledged that "automobiles enjoy a lesser expectation of privacy in our society than private homes," but we also noted that "citizens have a reasonable expectation of privacy from search and seizure in their cars and in their persons as they travel the state's roads." Id. at 672 (citing New York v. Class , 475 U.S. 106, 112, 106 S.Ct. 960, 89 L.Ed.2d 81 (1986) ("A citizen does not surrender all the protections of the Fourth Amendment by entering an automobile.")).

¶33 However, in People v. Esparza , we reviewed our prior precedent on dog sniffs and, in doing so, narrowed its scope. 2012 CO 22, 272 P.3d 367. We concluded that, "[d]espite our broad language purporting to address dog sniffs in general," our earlier decisions suggested a "more limited rule" was appropriate. Id. at ¶ 9, 272 P.3d at 369. Specifically, we noted that our dog-sniff jurisprudence has "been concerned with the nature of the particular container being pursued and any official conduct making it possible for a trained drug dog to sniff the container in the first place." Id. at ¶ 8, 272 P.3d at 369. Ultimately, we held "that an interest in possessing contraband cannot be deemed legitimate under the state constitution any more than under the federal constitution, and that official conduct failing to compromise any legitimate interest in privacy cannot be deemed a search under the state constitution any more than under the federal constitution." Id. at ¶ 2, 272 P.3d at 368. Like the U.S. Supreme Court, we based our holding in part on the purportedly binary nature of the dog's sniff. The sniff, we reasoned, could only communicate either the presence or absence of contraband. Id. at ¶ 11, 272 P.3d at 370.

¶34 Esparza thus represented a kind of return to the federal fold. The state of the law was thus: (1) under the automobile exception, an officer may search a car without a warrant if that officer has probable cause to believe there is contraband in the car; (2) an officer doesn't need reasonable suspicion to walk a drug-detection dog around a lawfully stopped vehicle because the alert indicates only the presence of contraband, and people do not have a legitimate expectation of privacy in contraband; and (3) an alert from a drug-detection dog could provide probable cause for a subsequent hand search of the car. See id. at ¶¶ 10-12, 272 P.3d at 370 ; see also People v. Mason , 2013 CO 32, ¶ 10, 310 P.3d 1003, 1005 ("It is now settled that walking a trained narcotics detection dog around a car that has not been unlawfully stopped or detained does not implicate the protections of either the Fourth Amendment or Article II, section 7 of the state constitution."). And so it has been for the better part of a decade.

D. Colorado Law: Post-Amendment 64

¶35 Has the passage of Amendment 64 altered this settled terrain? We began to explore this question in our recent decisions in People v. Zuniga , 2016 CO 52, 372 P.3d 1052, and People v. Cox , 2017 CO 8, 401 P.3d 509. In both Zuniga and Cox , we found probable cause supporting an automobile search based on a confluence of factors, including the positive alert of a drug-detection dog that was trained to alert to marijuana. Yet, in Zuniga , we concluded that the alert was *406legally ambiguous because a drug-detection dog can't distinguish legal marijuana from illegal marijuana, or legal marijuana from illegal narcotics. See ¶¶ 18-23, 372 P.3d at 1057-59. Despite this ambiguity, we held that the alert was still relevant to the overall probable cause analysis. Id. Likewise in Cox , we concluded that the positive alert of a drug-detection dog was one factor, among many, supporting a finding of probable cause to search a stopped vehicle. ¶ 17, 401 P.3d at 512-13.

¶36 In both Zuniga and Cox , we declined to address (1) whether the sniff of a dog trained to detect marijuana was a search, and (2) whether a positive alert from a dog trained to detect marijuana alone could establish probable cause. Significantly however, these two recent decisions suggest the answer to the latter question is no. We acknowledged that, with the legalization of small amounts of marijuana, a dog's alert doesn't provide a yes-or-no answer to the question of whether illegal narcotics are present in a vehicle. At most, the alert could be "suggestive of criminality," but not determinative on its own. See Zuniga , ¶ 23, 372 P.3d at 1059. Had the alert alone sufficed, there would have been no need to evaluate other circumstances in either Zuniga or Cox . Yet, we did just that. Thus, in finding that an alert from a drug-detection dog trained to find marijuana is ambiguous, Zuniga and Cox represent a tentative step away from the federal approach. See, e.g. , Harris , 568 U.S. at 240, 247-48, 133 S.Ct. 1050 (concluding that an alert from a well-trained, reliable drug-detection dog could supply probable cause to search a vehicle).

¶37 To be sure, marijuana remains illegal under federal law. The Federal Controlled Substances Act prohibits the possession of marijuana for nearly all uses. 21 U.S.C. § 812(c)(10) (2012) ; 21 U.S.C. § 812(b)(1)(A)-(C) (2012) ; 21 U.S.C. § 844(a) (2012). "There is no exception for marijuana use for medical purposes, nor is there an exception for use in compliance with state law." People v. Crouse , 2017 CO 5, ¶ 11, 388 P.3d 39, 42 (citing Gonzales v. Raich , 545 U.S. 1, 14, 125 S.Ct. 2195, 162 L.Ed.2d 1 (2005) ). Consequently, marijuana remains contraband under federal law and, arguably, Kilo's sniff was not a search under the U.S. Supreme Court's interpretation of the Fourth Amendment in Caballes .4 See McKnight , ¶ 18 n.4. This, however, leaves unanswered the question of whether Kilo's sniff was nonetheless a search under the corresponding provisions of our state constitution.5

¶38 We have opined that, despite the substantial similarity between article II, section 7 and the Fourth Amendment, "we are not bound by the United States Supreme Court's interpretation of the Fourth Amendment when determining the scope of state constitutional protections." Sporleder , 666 P.2d at 140. "There is no reason to think, as an interpretive matter, that constitutional guarantees of independent sovereigns, even guarantees with the same or similar words, must be construed in the same way." Jeffrey S. Sutton, 51 Imperfect Solutions: States and the Making of American Constitutional Law 174 (2018); see also *407Thompson v. Dallas City Attorney's Office , 913 F.3d 464, 471 (5th Cir. 2019) ("Our Nation boasts not one Constitution but 51, meaning American constitutionalism concerns far more than what began in Philadelphia 232 years ago." (citing Sutton, supra )). Indeed, the U.S. Supreme Court has a long tradition of encouraging us to scrutinize our own state constitutional texts, declaring: "It is fundamental that state courts be left free and unfettered by us in interpreting their state constitutions." Minnesota v. Nat'l Tea Co. , 309 U.S. 551, 557, 60 S.Ct. 676, 84 L.Ed. 920 (1940) ; see also Cooper v. California , 386 U.S. 58, 62, 87 S.Ct. 788, 17 L.Ed.2d 730 (1967) ("Our holding, of course, does not affect the State's power to impose higher standards on searches and seizures than required by the Federal Constitution if it chooses to do so."); Massachusetts v. Upton , 466 U.S. 727, 739, 104 S.Ct. 2085, 80 L.Ed.2d 721 (1984) (Stevens, J., concurring) ("The States in our federal system ... remain the primary guardian of the liberty of the people."); California v. Greenwood , 486 U.S. 35, 43, 108 S.Ct. 1625, 100 L.Ed.2d 30 (1988) ("Individual States may surely construe their own constitutions as imposing more stringent constraints on police conduct than does the Federal Constitution."). Thus, state courts need not interpret state constitutional provisions "as mere mirrors of federal protections." Developments in the Law-The Interpretation of State Constitutional Rights , 95 Harv. L. Rev. 1324, 1356 (1982) ; see also William J. Brennan, Jr., State Constitutions and the Protection of Individual Rights , 90 Harv. L. Rev. 489, 491 (1977) ("State constitutions, too, are a font of individual liberties, their protections often extending beyond those required by the Supreme Court's interpretation of federal law.").

¶39 And, because the protections afforded by both article II, section 7 and the Fourth Amendment are "highly generalized," we discern no reason to reflexively assume that there must be "just one meaning over a range of differently situated sovereigns." Sutton, supra , at 174. Criminal law has traditionally been considered best left to the expertise of the state courts as the vast majority of criminal prosecutions take place in state, rather than federal, court. See Shirley S. Abrahamson, Criminal Law and State Constitutions: The Emergence of State Constitutional Law , 63 Tex. L. Rev. 1141, 1150 (1985) ("Criminal law is an area of traditional concern for state judges. It is an area of law in which state judges have special experience and expertise."); see also Tafflin v. Levitt , 493 U.S. 455, 465, 110 S.Ct. 792, 107 L.Ed.2d 887 (1990) (reasoning, in part, that state courts can exercise concurrent jurisdiction over civil RICO cases because these cases "involve asserted violations of state law ... over which state courts presumably have greater expertise").

¶40 Moreover, we are confronted with a local development that further suggests resolution under our state constitution is proper. "State courts ... have a freer hand in doing something the Supreme Court cannot: allowing local conditions and traditions to affect their interpretation of a constitutional guarantee and the remedies imposed to implement that guarantee." Sutton, supra , at 17. When there are "general institutional differences between the state government and its federal counterpart" or "distinctive state-specific factors," including the existence of state constitutional provisions that provide rights not guaranteed by the federal constitution, interpreting the state constitution in a manner that departs from federal courts' interpretations of similar federal provisions not only makes sense, it may be the most logical option. See Developments in the Law , supra , at 1359-60.

¶41 Here, we have a state constitutional right not guaranteed by the federal constitution. Amendment 64 provides that it is "not unlawful and shall not be an offense under Colorado law" for a person who is at least twenty-one years of age to possess one ounce or less of marijuana. Colo. Const. art. XVIII, § 16 (3); see also Zuniga , ¶ 18, 372 P.3d at 1057. Amendment 64 was enacted through the initiative process in 2012, and its passage represents voters' intent to legalize marijuana use, possession, and growth under certain circumstances. Colo. Const. art. XVIII, § 16 (1)(a) (designating the purpose of Amendment 64 as "the people of the state of Colorado find and declare that the use of marijuana should be legal for persons twenty-one *408years of age or older and taxed in a manner similar to alcohol"). "Our goal in interpreting a citizen initiative is to 'give effect to the electorate's intent.' " People v. Lente , 2017 CO 74, ¶ 16, 406 P.3d 829, 832 (quoting Colo. Ethics Watch v. Senate Majority Fund, LLC , 2012 CO 12, ¶ 20, 269 P.3d 1248, 1253 ); see also Commonwealth v. Craan , 469 Mass. 24, 13 N.E.3d 569, 577-78 (2014) (discussing the importance of implementing the people's intent when examining the impact of a citizen initiative decriminalizing possession of small amounts of marijuana under state law on the probable cause analysis). Tellingly, Amendment 64 did more than just legalize personal marijuana use, possession, and growth. The initiative also created a marijuana industry. See Colo. Const. art. XVIII, § 16 (4). And, Amendment 64 reflects a desire to protect "individual privacy" by specifying that the Department of Revenue "shall not require" individual consumers to provide any information beyond a government-issued form of ID for age verification when purchasing marijuana at a retail marijuana store. Id. § 16 (5)(c).

¶42 Marijuana is not only decriminalized in Colorado, it is legalized, regulated, and taxed. Marijuana is now treated like guns, alcohol, and tobacco-while possession of these items is lawful under some circumstances, it remains unlawful under others. See, e.g. , § 18-12-103, C.R.S. (2018) (making it unlawful to knowingly possess a firearm without its serial number or other identifying mark); § 18-13-122(3)(a), C.R.S. (2018) (making it unlawful for persons under twenty-one to possess or consume alcohol); § 18-13-121, C.R.S. (2018) (making it unlawful to sell tobacco products to persons under eighteen). Although possession of guns, alcohol, and tobacco can be unlawful, persons still maintain an expectation of privacy in lawfully using or consuming those items. The same now goes for marijuana: In legalizing marijuana for adults twenty-one and older, Amendment 64 expanded the protections of article II, section 7 to provide a reasonable expectation of privacy to engage in the lawful activity of possessing marijuana in Colorado.

¶43 So, the rationale underlying Esparza no longer holds true when applied to Kilo's formidable nose. Because persons twenty-one or older may lawfully possess marijuana in small amounts, a drug-detection dog that alerts to even the slightest amount of marijuana can no longer be said to detect "only" contraband. Thus, an exploratory sniff of a car from a dog trained to alert to a substance that may be lawfully possessed violates a person's reasonable expectation of privacy in lawfully possessing that item. Because there was no way to know whether Kilo was alerting to lawful marijuana or unlawful contraband, Kilo's sniff violated McKnight's reasonable expectation of privacy. Therefore, under state law, Kilo's sniff was a search that had to be constitutionally justified.

¶44 Kilo's sniff now seems more akin to the "sense-enhancing" technology at issue in Kyllo . See 533 U.S. at 30, 121 S.Ct. 2038. There, the U.S. Supreme Court held that the use of a thermal-imaging device that was not used by the general public was a search because it could reveal private, lawful activity occurring inside of a home. Id. at 40, 121 S.Ct. 2038. With the device, law enforcement could "explore details ... that would previously have been unknowable without physical intrusion." Id. Though the U.S. Supreme Court focused its holding in part on the fact that the device was used on the home, Kyllo 's central rationale is instructive because "citizens [still] have a reasonable expectation of privacy from search and seizure in their cars and in their persons as they travel the state's roads." Haley , 41 P.3d at 672 ; accord Caballes , 543 U.S. at 413 n.3, 125 S.Ct. 834 (Souter, J., dissenting) ("Any difference between the dwelling in Kyllo and the trunk of the car ... may go to the issue of the reasonableness of the respective searches, but it has no bearing on the question of search or no search.... The justifications required by the Fourth Amendment may or may not differ as between the two practices, but if constitutional scrutiny is in order for the imager, it is in order for the dog.").

¶45 One might be tempted to underestimate the privacy interests at stake here. Consider how the Kyllo Court emphasized that a thermal imager could reveal "intimate details" of the home in reaching its conclusion that the use of the imager constituted a *409search. 533 U.S. at 38, 121 S.Ct. 2038. In contrast, in Dow Chemical Co. v. United States , the Court reasoned in part that aerial photographs of an industrial complex didn't violate the Fourth Amendment because the photos were "not so revealing of intimate details" and commercial property enjoys a lesser degree of privacy than the home. 476 U.S. 227, 237-38, 106 S.Ct. 1819, 90 L.Ed.2d 226 (1986). As the Court stated, "[t]he intimate activities associated with family privacy and the home and its curtilage simply do not reach the outdoor areas or spaces between structures and buildings of a manufacturing plant." Id. at 236, 106 S.Ct. 1819. Though automobiles have traditionally been accorded lesser protections than homes, a person surely engages in more "intimate activities" in the solitary confines of his car as compared to the "outdoor areas or spaces between buildings of a manufacturing plant." Id. Such intimate activities may include lawful transportation of an ounce or less of marijuana to treat symptoms of a chronic illness, mitigate mental health conditions, or engage in recreational activities that, while legal, one would just as soon not have disclosed to the public. See Leg. Council Colo. Gen. Ass., Research Publ'n No. 475-0, An Analysis of the 2000 Statewide Ballot Proposals 2 (2000); Leg. Council Colo. Gen. Ass., Research Publ'n No. 614, 2012 State Ballot Information Booklet 12 (2012). And Kilo's extraordinarily powerful nose can uncover this lawful activity. Though the sniff wouldn't necessarily result in the disclosure of why a person is carrying lawful amounts of marijuana, a person may still reasonably expect that his decision to engage in lawful activity for such personal reasons to remain private from others, including from a police officer and his canine.

¶46 We certainly aren't the first to notice the similarities between a well-trained drug-detection dog and other forms of investigative, sense-enhancing technologies that can reveal private, lawful activity. See, e.g. , Unruh , 713 P.2d at 376-77 (discussing the rationale of a minority of courts that concluded a sniff was a search). As the concurrence observed in United States v. Bronstein :

There is no legally significant difference between the use of an X-ray machine or magnetometer to invade a closed area in order to detect the presence of a metal pistol or knife, which we have held to be a search, and the use of a dog to sniff for marijuana inside a private bag. Each is a non-human means of detecting the contents of a closed area without physically entering into it.... Neither constitutes a particularly offensive intrusion, such as ransacking the contents of the hidden space, or exposing a person to indignities in the case of a personal search. But the fact remains that each detects hidden objects without actual entry and without the enhancement of human senses. The fact that the canine's search is more particularized and discriminate than that of the magnetometer is not a basis for a legal distinction. The important factor is not the relative accuracy of the sensing device but the fact of the intrusion into a closed area otherwise hidden from human view, which is the hallmark of any search.

521 F.2d 459, 464 (2d Cir. 1975) (Mansfield, J., concurring) (citation omitted); see also Caballes , 543 U.S. at 413, 125 S.Ct. 834 (Souter, J., dissenting) ("[T]he sniff is the first step in a process that may disclose 'intimate details' without revealing contraband, just as a thermal-imaging device might do, as described in Kyllo v. United States ."). And Kilo's sniff isn't all that different from the magnetometer or the thermal-imaging device. Kilo has spent countless hours being trained to detect marijuana-he can even alert to substances placed in a closed box and further sealed off with duct tape. Cf. Katz , 389 U.S. at 361-62, 88 S.Ct. 507 (Harlan, J., concurring) (discussing how the defendant exhibited a subjective expectation of privacy in his phone calls by entering a phone booth and shutting the door behind him, and how that expectation was reasonable). "[Drug-detection dogs] are to the poodle down the street as high-powered binoculars are to a piece of plain glass. Like the binoculars, a drug-detection dog is a specialized device for discovering objects not in plain view (or plain smell)." Jardines , 569 U.S. at 13, 133 S.Ct. 1409 (Kagan, J., concurring).

¶47 Admittedly, many federal courts have come to a different conclusion on whether a sniff of the exterior of a vehicle, or the air *410surrounding an item, constitutes a search. These courts have reasoned that a person doesn't have a legitimate expectation of privacy under the Fourth Amendment in the odorous molecules that emanate from his car into the open and public airspace. See, e.g. , United States v. Morales-Zamora , 914 F.2d 200, 205 (10th Cir. 1990) ; United States v. Broadway , 580 F. Supp. 2d 1179, 1191 (D. Colo. 2008). However, in those cases, the "odorous molecules" that a drug-detection dog could alert to all belonged to what was clearly contraband-that isn't true anymore in Colorado, where some odorous molecules may be those emanating from lawfully possessed marijuana.6 In short, none of these cases is on point because none involved legalized marijuana. And that is precisely why we rest our holding on the Colorado Constitution, not on the Fourth Amendment. To the extent we end up alone on a jurisprudential island, it is an island on which Colorado voters have deposited us. Our role is not to question their decision. Rather, it is to apply the logic of existing law to a changing world. Though we are the first court to opine on whether the sniff of a dog trained to detect marijuana in addition to other substances is a search under a state constitution in a state that has legalized marijuana, we probably won't be the last. Since Amendment 64's passage in 2012, nine other states and the District of Columbia have legalized possession of small amounts of marijuana. Marijuana Overview , Nat'l Conference of State Legislatures (Dec. 14, 2018), http://www.ncsl.org/research/civil-and-criminal-justice/marijuana-overview.aspx [https://perma.cc/HMN2-PX38].

¶48 Because a sniff from a dog trained to detect marijuana (in addition to other substances) can reveal lawful activity, we conclude that sniff is a search under article II, section 7 and must be justified by some degree of suspicion of criminal activity.

E. Probable Cause Was Necessary for Kilo's Search

¶49 Because we have concluded that the sniff from a dog trained to detect marijuana is a search under article II, section 7, it follows that the search must be justified by probable cause. However, as we are the first court to consider the impact of marijuana legalization on a sniff from a dog trained to alert to even a hint of marijuana, we consider whether reasonable suspicion would suffice, as the division below concluded after reviewing our pre- Esparza caselaw, or if the intrusion warrants justification under the usual probable cause standard.

¶50 "There is no more basic constitutional rule ... than that which makes a warrantless search unreasonable except in a few 'jealously and carefully drawn' exceptional circumstances." United States v. Watson , 423 U.S. 411, 427, 96 S.Ct. 820, 46 L.Ed.2d 598 (1976) (quoting Jones v. United States , 357 U.S. 493, 499, 78 S.Ct. 1253, 2 L.Ed.2d 1514 (1958) ). Because warrantless searches are presumptively unreasonable, they must be supported by at least probable cause and justified under at least one of these "carefully drawn" exceptions. Id. ; see also Chambers v. Maroney , 399 U.S. 42, 51, 90 S.Ct. 1975, 26 L.Ed.2d 419 (1970) ("In enforcing the Fourth Amendment's prohibition against unreasonable searches and seizures, the Court has insisted upon probable cause as a minimum requirement for a reasonable search permitted by the Constitution."). These two bedrock principles apply just as forcefully to the protections promised by article II, section 7. See Sporleder , 666 P.2d at 144 ("In the absence of some narrowly defined exception to the warrant requirement ... a warrantless search is presumed to be invalid."); Mendez v. People , 986 P.2d 275, 279 (Colo. 1999) ("[A] warrantless search is invalid unless it is supported *411by probable cause and is justified under one of the narrowly defined exceptions to the warrant requirement."). Our analysis thus begins with the presumption that a warrantless search must be supported by probable cause and justified under an exception to the warrant requirement.

¶51 "A police officer has probable cause to conduct a search when 'the facts available to [the officer] would warrant a [person] of reasonable caution in the belief' that contraband or evidence of a crime is present." Zuniga , ¶ 16, 372 P.3d at 1057 (alterations in original) (quoting Harris , 568 U.S. at 243, 133 S.Ct. 1050 ). "[P]robable cause is a commonsense concept that requires judges to consider the totality of the circumstances to determine 'whether a fair probability exists that a search of a particular place will reveal contraband or evidence of a crime.' " Id. (quoting Mendez , 986 P.2d at 280 ). This standard does not "lend itself to mathematical certainties," and, instead, "is based on factual and practical considerations of everyday life on which reasonable and prudent people, not legal technicians, act." Mendez , 986 P.2d at 280. But it is more demanding than reasonable suspicion. People v. Polander , 41 P.3d 698, 703 (Colo. 2001) ("Reasonable suspicion is a less demanding standard ... not only in the sense that it can be established with information that is different in quantity or content from that required for probable cause, but also in the sense that reasonable suspicion can arise from information that is less reliable than that required to show probable cause.").

¶52 Reasonable suspicion still requires "specific and articulable facts, greater than a mere hunch" to support a belief that a person was engaging in or had been engaged in criminal activity. People v. Boylan , 854 P.2d 807, 812 (Colo. 1993), abrogated by Esparza , 2012 CO 22, 272 P.3d 367. An officer can conduct a brief, investigatory stop if he has reasonable suspicion to believe that criminal activity "has occurred, is taking place, or is about to take place." People v. Ingram , 984 P.2d 597, 603 (Colo. 1999). And "[s]uspicion of even a minor traffic offense can provide the basis for a stop." People v. Chavez-Barragan , 2016 CO 16, ¶ 10, 365 P.3d 981, 983.

¶53 Two of the judges on the division below agreed that Kilo's sniff was a search requiring reasonable suspicion after looking back to the reasoning of our pre- Esparza caselaw. McKnight , ¶¶ 19-20 (opinion by Dailey, J.) (citing sources). In Unruh , we concluded that requiring reasonable suspicion to justify a dog-sniff search struck the proper "balance" between the government's interest in stopping the illegal narcotics trade and an individual's interest in being free from governmental search and seizure. 713 P.2d at 379. We reasoned that this was so because a dog's sniff was less intrusive than other forms of searches because it was "uniquely selective, detecting in most cases only the presence of narcotics." Id. If a dog alerted, in theory, it would only disclose to the officers the presence of contraband-the sniff couldn't disclose any specific details regarding the lawful contents of a vehicle. Accordingly, we concluded the dog-sniff search was minimally intrusive, and required only reasonable suspicion rather than probable cause to utilize. Id.

¶54 However, Unruh , and the line of cases following, were decided when marijuana was still wholly illegal under state law and, thus, a drug-detection dog's sniff could only uncover the presence or absence of illegal contraband. And perhaps at the time, treating a sniff from a drug-detection dog as a minimally invasive search only necessitating reasonable suspicion did balance the interests at stake.7 But that is no longer the case. The sniff can now reveal details concerning lawful activity, the possession of up to an ounce of marijuana by adults twenty-one and older. We see no reason to treat a search that could *412reveal legal marijuana any differently from a search that could reveal other lawful activity. Cf. Arizona v. Hicks , 480 U.S. 321, 328-29, 107 S.Ct. 1149, 94 L.Ed.2d 347 (1987) (refusing to create a sort of quasi-search that need only be justified by reasonable suspicion because it was less intrusive than other search tactics). Regardless of the relative intrusiveness or invasiveness, if law enforcement action constitutes a search, it must be authorized by a warrant from a neutral magistrate or justified by an exception to the warrant requirement and supported by probable cause.

¶55 Thus, we conclude that a sniff from a dog trained to alert to marijuana is a search in Colorado that must be supported by probable cause and justified under an exception to the warrant requirement, such as the automobile exception.

F. There Was No Probable Cause for Kilo's Search

¶56 So before Kilo's alert, what factors suggested there was probable cause to believe McKnight's car contained illegal narcotics under state law? The People emphasize that McKnight parked outside a house in which drugs had been found and that Officer Gonzales knew that McKnight's passenger had used methamphetamine in the past. But McKnight's decision to park near the house tells us little, considering that almost two months had passed since drugs were found there. See People v. Miller , 75 P.3d 1108, 1114-15 (Colo. 2003) (concluding that information about alleged criminal activity that was nearly a month old was too "stale" to provide probable cause for a search warrant). Moreover, even if we were to speculate (without record support) that the house was and remained a drug den, the record reveals that, on the evening in question, no one exited the truck or the house. Therefore, the People's observation that McKnight stopped his vehicle long enough to effectuate a drug deal and use any substance that had been bought is irrelevant. And we have consistently rejected the notion that "a history of past criminal activity in an area is itself sufficient to create a reasonable suspicion that a crime is being, has been, or will be committed." Outlaw v. People , 17 P.3d 150, 157 (Colo. 2001) (citing People v. Greer , 860 P.2d 528, 531 (Colo. 1993) ; People v. Rahming , 795 P.2d 1338, 1342 (Colo. 1990) ). If "a history of past criminal activity in an area" wasn't enough on its own to establish reasonable suspicion of criminal activity, then it also wouldn't be enough on its own to satisfy the more stringent standard of probable cause.

¶57 As for the passenger's drug history, Officer Gonzales could offer no details. All he could say with certainty is that whatever drug usage he recalled "wasn't super recent." The People encourage us to take notice of how difficult it is to overcome methamphetamine addiction and to extrapolate that the passenger was still using. On a record so lacking in details, we decline the invitation. Even if we accepted, it isn't clear that it would yield probable cause to believe that contraband would be found in the truck. Cf. Greer , 860 P.2d at 531 n.2 (citing Sibron v. New York , 392 U.S. 40, 62-63, 88 S.Ct. 1889, 20 L.Ed.2d 917 (1968) ("The inference that persons who talk to narcotics addicts are engaged in the criminal traffic in narcotics is simply not the sort of reasonable inference required to support an intrusion by the police upon an individual's personal security.")); People v. Feltch , 174 Colo. 383, 483 P.2d 1335, 1337 (1971) (rejecting "the theory that birds of a feather flock together" and stating "[g]uilt by association has never been an acceptable rationale and it does not constitute probable cause to arrest"). We are loathe to create a rule that would allow police to randomly search known drug users. Cf. Robinson v. California , 370 U.S. 660, 666-68, 82 S.Ct. 1417, 8 L.Ed.2d 758 (1962) (holding that a California statute that criminalized narcotic addiction itself was unconstitutional). In other words, drug users have the same constitutional rights as everyone else.

¶58 The question becomes: Are (1) proximity to a house in which drugs had been found nearly two months earlier, and (2) the presence of someone who "at some point" had used an illegal drug enough to establish a fair probability that contraband under state law would be found in McKnight's truck on the evening in question, thus justifying Kilo's sniff? Even when considered together, these *413two facts are insufficient to establish probable cause to believe McKnight's truck contained illegal narcotics under state law. We can look to Zuniga for an illuminating comparison. There we found probable cause to search a car because the driver and passenger had "divergent" stories about their visit to Colorado, they were "extreme[ly] nervous[ ]," there was a "strong odor of raw marijuana," and a drug-detection dog alerted at the back of the vehicle. Zuniga , ¶ 1, 372 P.3d at 1054. Though we did not need to resolve whether the sniff constituted a search in that case, we noted that an alert alone was ambiguous in light of Amendment 64. Id. at ¶¶ 20-21. And the totality of the circumstances-the "differing" stories, the "exceptional" nervousness, and the strong scent of unburnt marijuana that the officer was able to smell with his own nose-supported the officer's conclusion that there was a fair probability that a search would reveal evidence of a crime. Id. Here though, without Kilo's alert, the officers just observed McKnight's truck near a house where the last documented criminal activity occurred nearly two months before and McKnight's passenger was a person who had used illegal drugs at some point in the not "super recent" past. That isn't enough to warrant the sniff, or the subsequent intrusion into McKnight's truck. See Greer , 860 P.2d at 531 & n.2.

¶59 Thanks to the subsequent hand search of McKnight's truck, which only revealed the pipe with methamphetamine residue, we can infer that Kilo was not alerting to the scent of lawfully possessed marijuana. But, as noted at the outset of our analysis, we don't measure a search's constitutionality by looking at what it reveals. Rather, we examine the constitutionality of a search based on what was known at the time it was conducted.

¶60 And, like the division, we cannot conclude that "the evidence properly received against [McKnight was] so overwhelming that the constitutional violation was harmless beyond a reasonable doubt." McKnight , ¶ 24 (opinion of Dailey, J.) (citing Bartley v. People , 817 P.2d 1029, 1034 (Colo. 1991) ). Indeed, the People do not contend otherwise. Opening Brief at 10 ("The People concede that if this Court [were] to find that the trial court erred in denying McKnight's suppression motion, the error could not be harmless.").

G. Exclusion of the Pipe Is the Remedy for this Violation of the Colorado Constitution

¶61 Exclusion is the appropriate remedy for this constitutional violation. Though we have previously affirmed suppression orders excluding evidence obtained in violation of article II, section 7, we have not explicitly addressed whether we adopted a state corollary to the federal exclusionary rule. See, e.g. , Oates , 698 P.2d at 816, 819 (Colo. 1985) (affirming the suppression of evidence obtained from an illegal search under article II, section 7, without expressly analyzing exclusion as the remedy). Yet, in affirming suppression of evidence obtained in violation of article II, section 7, we have implied that we are open to finding exclusion to be the proper remedy for violations of state constitutional law. See Corr , 682 P.2d at 22, 27-28 (rejecting the application of the statutory good faith exception, and in affirming the suppression of toll telephone call records obtained in violation of article II, section 7, suggesting we approved of exclusion as the remedy); see also 1 Wayne R. LaFave, Search & Seizure: A Treatise on the Fourth Amendment § 1.5(b) (5th ed. Oct. 2018 update) (noting that state courts often suppress evidence obtained in violation of state constitutional prohibitions against unreasonable searches and seizures while "seldom" explaining their rationale for concluding "that the exclusionary remedy is just as appropriate [for state law violations] as when the Fourth Amendment is violated"). To the extent that we have not yet explicitly embraced the exclusionary rule as a remedy for violations of article II, section 7, we do so now. We recognize that the federal courts have created several exceptions to the exclusionary rule, see, e.g. , United States v. Leon , 468 U.S. 897, 104 S.Ct. 3405, 82 L.Ed.2d 677 (1984), and that we have our own statutory good-faith exception, see § 16-3-308, C.R.S. (2018), but we need not address the applicability of those exceptions. Because the prosecution did not raise any arguments about the exclusionary rule or its exceptions at the *414suppression hearing, we consider those arguments waived and decline to address them on appeal. See People v. Crippen , 223 P.3d 114, 116-17 (Colo. 2010) ; People v. Donahue , 750 P.2d 921, 923 (Colo. 1988).

III. Conclusion

¶62 We hold that a sniff from a drug-detection dog that is trained to alert to marijuana constitutes a search under article II, section 7 of the Colorado Constitution because that sniff can detect lawful activity, namely the legal possession of up to one ounce of marijuana by adults at least twenty-one years old. We further hold that, in Colorado, law enforcement officers must have probable cause to believe that an item or area contains a drug in violation of state law before deploying a drug-detection dog that alerts to marijuana for an exploratory sniff. Because there was no such probable cause justifying Kilo's search of McKnight's truck, the trial court erred in denying McKnight's motion to suppress.

¶63 This holding is based solely on the protections afforded by our state constitution, and to the extent we discuss any federal cases in coming to our conclusion, it is only for the purpose of analogy and comparison. See Michigan v. Long , 463 U.S. 1032, 1040-41, 103 S.Ct. 3469, 77 L.Ed.2d 1201 (1983).

¶64 Accordingly, we affirm the court of appeals' decision to reverse McKnight's judgment of conviction.

CHIEF JUSTICE COATS dissents, and JUSTICE BOATRIGHT and JUSTICE SAMOUR join in the dissent.

JUSTICE SAMOUR dissents, and JUSTICE BOATRIGHT joins in the dissent.

¶65 Apart from my objections to much of the majority's selective and at times revisionist promenade through the history of both federal and state search and seizure law, I object specifically to its radical reconstruction, in the wake of our recent marijuana initiative, of the state's own constitutional Bill of Rights. Although the majority does its best to avoid addressing the issue, I believe the most significant aspect of its rationale today concerns the question of federal supremacy, the nature and scope of which may be more at issue today, in various contexts, than has been the case at any time in the last hundred and fifty years. Because I not only disagree with the majority's outcome here but also consider its reasoning both deeply flawed and likely to have implications far beyond this case, I respectfully dissent and briefly outline the more significant of my objections.

¶66 I must begin, however, by registering my emphatic objection to the majority's gratuitous and wholly unjustified assertion that our prior decisions had already effectively foreclosed the possibility that the alert of a trained drug detection dog alone can any longer provide probable cause of the presence of contraband and therefore probable cause of the commission of a crime. In both People v. Zuniga , 2016 CO 52, ¶ 30 n.6, 372 P.3d 1052, 1060 n.6, and People v. Cox , 2017 CO 8, ¶ 22 n.5, 401 P.3d 509, 514 n.5, we expressly reserved the question, finding its resolution unnecessary in those cases, where there were other grounds establishing probable cause to search in addition to or in conjunction with the dog alert. The majority's suggestion that an appellate court's choice among alternate rationales to resolve an assignment of error implicitly indicates that all other possible rationales would be inadequate bespeaks a fundamental misconception about appellate practice and precedent, if not simply language and logic. In any event, the majority rationale today is a far cry from simply finding that a dog's inability to distinguish among quantities of marijuana deprives its alert of the capability, in and of itself, of establishing probable cause to believe that more than an ounce of the drug is present.

¶67 Lest the wheat of the real issue of the case be lost amongst the chaff, I emphasize the question upon which the majority actually resolves this case, which I understand to be whether the state constitution has now not only legalized the possession of small amounts of marijuana by adults in the state but has in fact shrouded such possession with an expectation of privacy such that its detection by police investigation within the *415territorial boundaries of this state amounts to the detection of "lawful" activity, despite the proscription of the very same activity as "unlawful" according to the coordinate federal criminal law. It seems clear that even the majority does not dispute the well-established proposition that the use of investigative techniques or devices, in this case a dog, capable of observing nothing more than activity that is unlawful, does not constitute a search within the contemplation of either the Fourth Amendment of the federal constitution or article II, section 7 of the state constitution. See Illinois v. Caballes , 543 U.S. 405, 408-09, 125 S.Ct. 834, 160 L.Ed.2d 842 (2005) ; People v. Esparza , 2012 CO 22, ¶ 11, 272 P.3d 367, 370. The specific wrinkle posed by our recent, popularly-initiated amendment to the state constitution is whether a dog sniff, despite being capable of exposing no more than the presence of federal contraband, nevertheless constitutes a search within the contemplation of the state constitution for the reason that it can detect the presence of a small amount of marijuana possessed by an adult, which is now lawful under state law.

¶68 In Coats v. Dish Network, LLC , 2015 CO 44, ¶ 20, 350 P.3d 849, 853, we addressed the similar question whether the General Assembly intended to extend the protections for "lawful" activities afforded by section 24-34-402.5, C.R.S. (2018), to activities made lawful under state law but remaining unlawful under federal law. We there concluded that the term "lawful" could not have been intended to include the possession of marijuana that remains "unlawful" under federal law, Coats , ¶ 20, 350 P.3d at 853, which in our federal system not only applies in this jurisdiction but, in the event of conflict, actually reigns supreme over state law. Today, the majority holds instead that a dog sniff capable of exposing the lawful presence of marijuana under state law, but no more than the possession of contraband-unlawful activity-under federal law, nevertheless constitutes a search within the contemplation of the state constitution, requiring probable cause and either a warrant or an exception thereto.

¶69 In the last half-century, the Supreme Court has come to construe the language of the Fourth Amendment protecting the "right of people to be secure in their persons, houses, papers, and effects against unreasonable searches and seizures," as intending to broadly protect any reasonable expectation of privacy whatsoever, see Katz v. United States , 389 U.S. 347, 361, 88 S.Ct. 507, 19 L.Ed.2d 576 (1967) (Harlan, J., concurring), which concept it has subsequently defined to include any subjective expectation of privacy that society is prepared to recognize as reasonable, California v. Ciraolo , 476 U.S. 207, 211, 106 S.Ct. 1809, 90 L.Ed.2d 210 (1986). Although without elaboration, on several occasions in the past, we apparently not only subscribed to this broad construction of the words "persons, houses, papers, and effects," by understanding article II, section 7 as similarly protecting any reasonable expectation of privacy but, in addition, accepted as reasonable certain expectations considered unreasonable by the Supreme Court. See Charnes v. DiGiacomo , 200 Colo. 94, 612 P.2d 1117, 1121 (1980) (finding reasonable expectation of privacy and therefore protection under the state constitution for records in the hands of banks); see also People v. Oates , 698 P.2d 811, 818-19 (Colo. 1985) (same for purchase of drug precursors); People v. Corr , 682 P.2d 20, 27 (Colo. 1984) (same for phone records); People v. Sporleder , 666 P.2d 135, 144 (Colo. 1983) (same for numbers of incoming phone calls detected by pen register). Today, following its paean to separate reliance on state constitutions, the majority construes our constitution, in conjunction with the regulatory framework created for its implementation, as not only decriminalizing the possession of small amounts of marijuana but also as reflecting a state policy adopting the proposition that an adult's subjective expectation of privacy in his possession of an amount of marijuana legal under state law must be considered objectively reasonable, with the effect that in Colorado such an individual is held to have a reasonable expectation of privacy in the commission of federal crime. Not only do I not share this view of state law as designating such a subjective expectation of privacy reasonable, but perhaps even more importantly, I do not share the majority's view that the marijuana amendments and regulations upon which it relies were intended to alter our *416prior interpretation of article II, section 7 with regard to the use of investigative techniques capable of detecting no more than unlawful activity.

¶70 Initially, I consider thoroughly unpersuasive the majority's patchwork of conclusory propositions in support of its conclusion that Amendment 64 not only legalized certain activities related to marijuana but in fact "expanded the protections of article II, section 7 to provide a reasonable expectation of privacy to engage in the lawful activity of possessing marijuana in Colorado." Maj. op. ¶ 42. The amendment, of course, does not cross-reference the constitutional search and seizure provision at all; does not in any way allude to an expectation of privacy in the possession of marijuana, much less a reasonable expectation of privacy in violating federal law; and does not remotely suggest the imposition of any restriction on state executive branch officers from detecting federal crime. Furthermore, the sole reference to "individual privacy" mentioned by the majority is limited to an express prohibition against the promulgation of regulations mandating the collection of personal data by retailers as a precondition to sale. Maj. op. ¶ 41.

¶71 Even if it were plausible to construe our regulatory scheme as intending to create a special expectation of privacy in the possession of marijuana, as does the majority, however, I can perceive nothing (and the majority offers nothing) in that construction requiring us to alter the calculus governing our decision in Coats . Just as in Coats , ¶ 20, 350 P.3d at 853, the question upon which the majority's ultimate holding turns is whether the possession of marijuana can be considered "lawful" activity in this context, despite its continued proscription as criminal by federal law-law which also governs this jurisdiction. In this regard, I consider the majority's reliance on a separate state constitutional expectation of privacy in the possession of marijuana to be superfluous, in no way rendering the possession of federal contraband in the state "lawful."

¶72 In particular, the majority notes a line of cases decided more than thirty years ago in which we held as a matter of state constitutional law that an individual does not abandon a reasonable expectation of privacy simply by exposing otherwise private information to third parties no more than is necessary to participate and conduct business in the modern world. See Oates , 698 P.2d at 818-19 (exposing the purchase of drug precursors to the chemical wholesaler); Corr , 682 P.2d at 26-27 (exposing numbers of outgoing calls to phone company); Sporleder , 666 P.2d at 144 (exposing numbers of incoming calls to phone company); DiGiacomo , 612 P.2d at 1121 (exposing banking transactions to the bank). While there can be no question that states may, and when appropriate should, afford even greater protections to their citizenry than those already afforded by the federal Bill of Rights, see Curious Theatre Co. v. Colo. Dep't of Pub. Health & Env't , 220 P.3d 544, 551 (Colo. 2009), by the same token states are clearly not empowered to enact laws that conflict with federal law, see Wyeth v. Levine , 555 U.S. 555, 563, 129 S.Ct. 1187, 173 L.Ed.2d 51 (2009). Nor have we ever suggested anything of the sort. Unless the majority intends to equate affording protections beyond those rights already afforded under the federal Bill of Rights with affording protections for violating federal law, I fail to see how its discussion of our past reliance on state constitutional grounds in any way advances its analysis. Unlike the majority, I do not believe the language of the marijuana initiative, even when considered in conjunction with other constitutional or statutory provisions, can be understood to recognize a reasonable expectation of privacy in committing federal crime, and I consider it clear that this court has never, for obvious reasons, interpreted article II, section 7 to sanction any such expectation.

¶73 Beyond extending in this way, without acknowledgement or justification of any kind, the effect of reliance on state constitutional grounds, the majority also overturns long-accepted precedent of the jurisdiction, similarly without acknowledgement or justification of any kind, by announcing, ex cathedra , that we now will "embrace[ ] the exclusionary rule as a remedy for violations of article II, section 7." Maj. op. ¶ 61. Although, as the *417majority indicates, we have on several occasions in the past affirmed suppression orders of lower courts as the result of violations of expectations of privacy recognized as reasonable solely under the state constitution, see Oates , 698 P.2d at 812 ; Corr , 682 P.2d at 22 ; Sporleder , 666 P.2d at 136, we have never suggested the existence of a separate exclusionary rule for violation of the state constitution. Whether the absence of any justification for adopting the suppression orders in those cases reflects an assumption that the exclusionary remedy adopted by the Supreme Court would require the suppression of an unjustified infringement on any reasonable expectation of privacy, regardless of its source, or reflects simply the absence of any express challenge to the affirmance of the lower courts' exclusionary orders, it is clearly substantially more problematic, even to the majority, to order the exclusion of evidence for violation of a state-recognized expectation of privacy in committing federal crime, without expressly resting that remedy in state law.

¶74 The last time this court expressly considered the question whether our constitutional guarantee against unreasonable searches implied its own exclusionary remedy, we firmly rejected that proposition. See Wolf v. People , 117 Colo. 279, 187 P.2d 926, 927-28 (1947), aff'd , Wolf v. Colorado , 338 U.S. 25, 69 S.Ct. 1359, 93 L.Ed. 1782 (1949), overruled by Mapp v. Ohio , 367 U.S. 643, 81 S.Ct. 1684, 6 L.Ed.2d 1081 (1961) (concerning the applicability in state criminal proceedings of the exclusionary rule adopted in Weeks v. United States , 232 U.S. 383, 34 S.Ct. 341, 58 L.Ed. 652 (1914), for evidence seized in violation of the Fourth Amendment). While we expressly rejected any exclusionary rule of evidence for violation of article II, section 7 in Wolf , this was not the only or first time we did so, see Roberts v. People , 78 Colo. 555, 243 P. 544, 559 (1926); Massantonio v. People , 77 Colo. 392, 236 P. 1019, 1021 (1925), and we continued to reject the exclusion of evidence for violation of constitutional search provisions until ordered to apply the Fourth Amendment exclusionary rule in Mapp , see Williams v. People , 136 Colo. 164, 315 P.2d 189, 191 (1957).

¶75 This court first rejected exclusion as a remedy for a violation of article II, section 7 in Massantonio , 236 P. at 1021, largely reasoning that the rule in Weeks was fundamentally flawed because the Fourth Amendment violation occurs when the illegal seizure happens, not when the evidence is used at trial; because there is nothing about the constitutional violation itself that renders the evidence incompetent; and because the rule contradicted the fundamental principle accepted at law that an illegality in the mode of procuring evidence is no ground for excluding it. We reaffirmed Massantonio in Roberts , 243 P. at 559, noting that the exclusionary rule had been rejected "after a thorough examination of all the available authorities and a careful consideration of the alleged reasons supporting each." Roberts also noted that, in addition to the logical errors made by the rule, it also lacked policy justification because it could, for instance, allow "one unquestionably guilty of an atrocious murder" to be "turned loose." Id.

¶76 In the absence of offering any rationale or justification for reversing course and now adopting an exclusionary remedy for violation of article II, section 7, the majority necessarily offers no hint about the nature and applicability of its new rule, beyond the fact that it will require the suppression of the meth pipe discovered in this case. Even the federal exclusionary rule itself is no longer a bright-line rule of automatic exclusion, having now been declared "a judicially created remedy designed to safeguard Fourth Amendment rights generally through its deterrent effect, rather than a personal constitutional right of the party aggrieved." United States v. Calandra , 414 U.S. 338, 348, 94 S.Ct. 613, 38 L.Ed.2d 561 (1974). In the ensuing years, the Supreme Court has expressly recognized a host of limitations and exceptions to exclusion, including the inapplicability of the rule in whole classes of proceedings, see Stone v. Powell , 428 U.S. 465, 494, 96 S.Ct. 3037, 49 L.Ed.2d 1067 (1976) (habeas corpus proceedings); United States v. Janis , 428 U.S. 433, 459-60, 96 S.Ct. 3021, 49 L.Ed.2d 1046 (1976) (civil proceedings); Calandra , 414 U.S. at 354, 94 S.Ct. 613 (grand jury proceedings); its inapplicability to the fruits of an unlawful search under *418various circumstances, see Nix v. Williams , 467 U.S. 431, 444, 104 S.Ct. 2501, 81 L.Ed.2d 377 (1984) (inevitable discovery); Silverthorne Lumber Co. v. United States , 251 U.S. 385, 392, 40 S.Ct. 182, 64 L.Ed. 319 (1920) (independent source); and even its ineffectiveness where otherwise applicable, see Herring v. United States , 555 U.S. 135, 137, 147, 129 S.Ct. 695, 172 L.Ed.2d 496 (2009) (negligence as to accuracy of active warrants); Hudson v. Michigan , 547 U.S. 586, 594, 126 S.Ct. 2159, 165 L.Ed.2d 56 (2006) (violations of knock-and-announce rule); Arizona v. Evans , 514 U.S. 1, 12-14, 115 S.Ct. 1185, 131 L.Ed.2d 34 (1995) (clerical errors); United States v. Leon , 468 U.S. 897, 918-21, 104 S.Ct. 3405, 82 L.Ed.2d 677 (1984) (good faith execution of defective warrants). In addition, the Supreme Court has made clear that even where no specific exception has been recognized, the rule may not apply if its potential deterrent benefits do not merit its costs. Herring , 555 U.S. at 140, 129 S.Ct. 695 (citing Hudson , 547 U.S. at 591, 126 S.Ct. 2159 ). Whether the new Colorado rule will follow the continuing evolution of the federal rule, or whether it will simply become another vehicle for disagreeing with the Supreme Court or providing insulation from its review, for now must apparently remain a mystery.

¶77 The legal justification for, or even policy benefits of, adopting an exclusionary rule as a means of modifying executive branch behavior having no impact on the competence of the evidence itself is hardly self-evident or a matter to be casually decided without vigorous debate. Both sides of the political spectrum have argued against the use of the federal exclusionary rule, despite its having been in existence for more than a century. See Christopher Slobogin, Why Liberals Should Chuck the Exclusionary Rule , 1999 U. Ill. L. Rev. 363, 365-66 (1999) (arguing that exclusion is appropriate only for flagrant violations and that monetary liability for individual officers and police departments would better deter Fourth Amendment violations); Patrick Tinsley, N. Stephan Kinsella & Walter Block, In Defense of Evidence and Against the Exclusionary Rule: A Libertarian Approach , 32 S.U. L. Rev. 63, 72 (2004) (arguing that removing exclusionary rule and instead allowing for damages suits would be more effective and would allow non-criminal victims of Fourth Amendment violations to have a remedy); see also Janis , 428 U.S. at 448-53, 96 S.Ct. 3021 (discussing difficulties in evaluating deterrence).

¶78 In addition, a number of substantial arguments have arisen specifically against the adoption of separate state constitutional exclusionary rules. See, e.g. , Paul G. Cassell, The Mysterious Creation of Search and Seizure Exclusionary Rules Under State Constitutions: The Utah Example , 1993 Utah L. Rev. 751, 756-58 (1993) ; Robert M. Pitler, Independent State Search and Seizure Constitutionalism: The New York State Court of Appeals' Quest for Principled Decisionmaking , 62 Brook. L. Rev. 1, 174-80 (1996). Pitler articulates a number of objections to a judicially created exclusionary rule by New York's highest court, including 1) various arguments raised against such a rule at one of New York's constitutional conventions, 2) precedent refusing to recognize an exclusionary rule, 3) common law evidentiary principles that reliable evidence should be admitted to further the truth-finding process, and 4) the inherently policy-based nature of the rule. Pitler, supra , at 174-75. Focusing on Utah's judicially created exclusionary rule, Cassell also provides a useful framework from which courts might analyze the question whether exclusion is an appropriate remedy under a state constitution, arguing essentially that state courts are obliged to begin by considering the intent of the framers of their own constitution's search and seizure provision, through traditional constitutional precepts of interpretation. Cassell, supra , at 757. Equally applicable to the question in this jurisdiction is his admonition that state courts must also consider whether an exclusionary rule is consistent with the state's separation of powers doctrine and rules of evidence, as well as whether such a rule would be sound from a state public policy perspective. Id. Finally, the availability and effectiveness of other remedies for search and seizure violations cannot be discounted in any meaningful assessment. Id. at 758.

¶79 Similarly, state high courts, including this court in particular, have been cautioned in adopting an exclusionary rule to avoid *419tying themselves too tightly to the federal rule as applied by the Supreme Court, given the haphazard and inconsistent evolution of that rule. See Michael K. Schneider, An Exclusionary Rule Colorado Can Call Its Own , 63 U. Colo. L. Rev. 207, 208 (1992) ; see also Jeffrey S. Sutton, 51 Imperfect Solutions: States and the Making of American Constitutional Law 42-83 (2018) (recounting the development of the federal exclusionary rule and encouraging state courts not to assume federal courts can best protect the constitutional rights of criminal defendants). Having done so itself, at least one state high court cautions that whatever process a state court uses to adopt or reject a state exclusionary rule, at the very least it should consider basic questions: 1) What is the basis of the rule-is it a constitutional right, merely a judicial remedy, or constitutionally compelled?; 2) What are the purposes behind the rule-deterrence alone or something else?; and 3) Who is meant to be deterred-police only or other persons in the criminal justice system? See State v. Larocco , 794 P.2d 460, 473 (Utah 1990) (plurality opinion) (adopting state exclusionary rule); see also Sanford E. Pitler, The Origin and Development of Washington's Independent Exclusionary Rule: Constitutional Right and Constitutionally Compelled Remedy , 61 Wash. L. Rev. 459, 460 (1986) (finding that state courts typically justify an exclusionary rule of evidence as a separate constitutional right, as a judicially created remedy, or as compelled by the applicable constitutional provision).

¶80 Beyond what I consider to be its logical flaws, legal misinterpretations, and intemperate extensions of state constitutional doctrine, however, I am particularly concerned that in going to such lengths to craft a rationale for imposing limitations on the use of drug detection dogs, the majority unwittingly exposes not only the marijuana initiative itself but even the state's constitutional Bill of Rights to a much greater risk of federal preemption than would previously have been the case. Whether or not the legalization and regulation of marijuana and the majority's construction of the state's Bill of Rights in the wake of those actions are of a piece, such that the majority's resolution today could in any way have been anticipated, I do not consider it appropriate, or wise, to sidestep what I believe to be the most significant impediment to the majority's rationale. While resting a decision solely on state grounds may generally be an effective technique for insulating state courts from further federal review, that is clearly not the case with regard to the supremacy of federal law.

¶81 Any federal system of government must of course have some strategy for dealing with conflicts between the national and local governments, and the framers of our federal constitution embodied their choice of strategy in the Supremacy Clause, U.S. Const. art. VI, cl. 2. Although the rules developed by the Supreme Court for determining the existence of an irreconcilable conflict requiring resolution in favor of federal law may be complex and at times unclear, even before today's construction we had found at least a portion of Amendment 64 to be preempted by federal law. See People v. Crouse , 2017 CO 5, ¶ 19, 388 P.3d 39, 43. Whether the recent marijuana law, on its face or as construed to impact article II, section 7, "stands as an obstacle to the accomplishment and execution of the full purposes and objectives of Congress," Hines v. Davidowitz , 312 U.S. 52, 67, 61 S.Ct. 399, 85 L.Ed. 581 (1941) ; see also People v. Crouse , 2013 COA 174, ¶ 91, 412 P.3d 599, 616 (Bernard, J., concurring in part and dissenting in part), rev'd , 2017 CO 5, 388 P.3d 39, appears to be a question that will have to wait for another day, but I believe that after today, that question cries out for resolution more than ever before.

¶82 Because I believe that a straightforward application of Zuniga and Esparza require that the trial court's denial of the motion to suppress be affirmed, I would reverse the judgment of the court of appeals.

¶83 I therefore respectfully dissent.

I am authorized to state that JUSTICE BOATRIGHT and JUSTICE SAMOUR join in this dissent.

¶84 Today the majority concludes that a dog sniff of the public air outside a car, *420though not a search under the Federal Constitution, is a search under the Colorado Constitution because Amendment 64 legalized the possession of marijuana in limited circumstances. However, in doing so, the majority skirts a critical question in the analysis: Does a driver have a reasonable expectation of privacy in the odors that escape from his car and become part of the public airspace? Instead, the majority: (1) draws an unwarranted inference from Illinois v. Caballes , 543 U.S. 405, 125 S.Ct. 834, 160 L.Ed.2d 842 (2005) ; and (2) analogizes a dog sniff of the public air outside a lawfully stopped car to the use of a thermal-imaging device aimed at a private home to detect relative amounts of heat inside. And, to boot, the majority purports to nestle its ruling in the Colorado Constitution when it is clearly anchored to Fourth Amendment jurisprudence. Accordingly, while I wholeheartedly join the Chief Justice's well-reasoned dissent, I write separately to express additional bases of disagreement with the majority's decision.

I. Overview

¶85 The majority's flawed reasoning goes something like this:

(1) the Fourth Amendment protects people from unreasonable searches and seizures, and warrantless searches are presumed unreasonable and, therefore, unconstitutional;

(2) the U.S. Supreme Court held in Katz v. United States , 389 U.S. 347, 88 S.Ct. 507, 19 L.Ed.2d 576 (1967), that a search under the Fourth Amendment occurs when the government violates a subjective expectation of privacy that society recognizes as reasonable;

(3) Amendment 64 created a reasonable expectation of privacy in the possession of an ounce or less of marijuana by adults twenty-one and older in Colorado;

(4) McKnight was older than twenty-one when his car was lawfully stopped;

(5) therefore, officers violated McKnight's rights under the Colorado Constitution (article II, section 7 ) when they had Kilo-a dog trained to detect the odor of drugs (including marijuana)-sniff the public air around his car without a warrant.

¶86 Even assuming Amendment 64 created the reasonable expectation of privacy the majority says it did, the majority's analysis is incomplete. To be sure, under the U.S. Supreme Court case law on which the majority opinion rests, McKnight had a reasonable expectation of privacy in any noncontraband items in his car. (Although I agree with the Chief Justice's views, I assume for purposes of this dissent that possession of up to an ounce of marijuana by someone who is at least twenty-one qualifies as possession of noncontraband in Colorado after the passage of Amendment 64). Even so, that leaves a question still to be answered: Did McKnight also have a reasonable expectation of privacy (under the Colorado Constitution or any other authority) in odors that emanated from noncontraband items in his car and escaped into the infinite public air outside? The majority circumvents this question, overlooking a focal point of the relevant inquiry: the nature of the place subjected to the dog sniff. After all, whether a dog sniff violates someone's reasonable expectation of privacy depends in large part on the place sniffed-be it a safe in a home, luggage at an airport, a trash can in a public park, a laundry bag in a correctional facility, or the air outside a car during a lawful traffic stop. If focusing solely on the privacy interest in the item that an officer's investigative technique (here, a dog sniff) is aimed at detecting or locating (here, drugs, including marijuana) were sufficient, it would lead to absurd results.

¶87 The majority relies heavily on the U.S. Supreme Court's decision in Caballes (and on our adoption of its rationale in People v. Esparza , 2012 CO 22, 272 P.3d 367 ). But, while Caballes dealt with a dog sniff of the exterior of a car during a lawful traffic stop, it did not have to resolve the question that is front and center in this case because marijuana was illegal in Illinois when Caballes was decided. It sufficed to hold in Caballes that no search had occurred because the dog sniff could have detected only the presence or absence of contraband, and no person can ever have a legitimate expectation of privacy in contraband. That approach cannot dispose of this appeal. The majority nevertheless assumes *421that, since the Court held in Caballes that the dog sniff was not a search because it could only detect contraband, the Court must have also meant that a dog sniff like the one here, which could detect a drug the majority now views as noncontraband, is a search.

¶88 In place of meaningful analysis regarding whether McKnight had a reasonable expectation of privacy in the public air sniffed by Kilo, the majority analogizes Kilo's sniff to the use of a thermal imager that measures from the street the amount of heat inside a private home, see Kyllo v. United States , 533 U.S. 27, 121 S.Ct. 2038, 150 L.Ed.2d 94 (2001). The failure to accord these drastically distinct factual scenarios the different treatment they deserve under the Fourth Amendment and article II, section 7 leads the majority astray, eventually declaring that the dog sniff here is as much a search under Katz as the thermal imaging of the home in Kyllo . Kyllo is inapposite for two reasons: (1) a thermal-imaging scanner that detects infrared radiation and converts it into images based on relative warmth can never be confused with a dog's nose; and (2) there is a night-and-day difference between a home and a car.

¶89 To top it all off, the majority seeks cover under the cloak of the Colorado Constitution. But the maneuver fails. While the legalization of marijuana pursuant to Amendment 64 is clearly Colorado-specific, the majority's search-and-seizure analysis is not; rather, it is clearly predicated on U.S. Supreme Court precedent interpreting the Fourth Amendment. The majority does not shepherd a new Colorado-only methodology for determining what constitutes a search. To the contrary, it recognizes that the applicable test is the one ushered in by the Court in Katz , it fastens its rationale to its reading of Caballes , and it analogizes this case to Kyllo . Thus, what is unique to Colorado in this case is the creation of an alleged new privacy interest in the possession of marijuana in limited circumstances-a point I do not contest for purposes of this dissent-not the adoption of a new analytical framework in the search-and-seizure arena or the marshaling of a new constitutional scheme to protect against unreasonable searches and seizures.

¶90 Using the Colorado Constitution as a broom, however, the majority attempts to sweep under the rug the overwhelming authority from other jurisdictions directly contravening its decision. Almost without exception, state and federal courts have held that a driver has no reasonable expectation of privacy in odors that escape from his car and are detected during a dog sniff of the public air outside. And no court has gone against that grain for at least thirty years. In concluding that Kilo's sniff was a search, the majority bucks the trend and places this court on an island. In response, the majority says that this island is one "on which Colorado voters have deposited us," and that "[o]ur role is not to question their decision." Maj. op. ¶ 47. But the finger should not be pointed at Colorado's voters. They were not asked about search-and-seizure issues, much less about dog sniffs, when they voted for Amendment 64.

¶91 Respectfully, I cannot join my colleagues in the majority. Instead, I would conclude that, while a driver in Colorado has a reasonable expectation of privacy in any noncontraband items and "lawful activity" in his car, id. at ¶ 3, he does not have a reasonable expectation of privacy in odors that emanate from those items or activities and escape into the infinite public air outside. Because McKnight had no reasonable expectation of privacy in the public air sniffed by Kilo, Kilo's sniff cannot be deemed a search under the Fourth Amendment or article II, section 7. Thus, even if, as the majority concludes, Amendment 64 created a reasonable expectation of privacy in the possession of up to an ounce of marijuana by anyone who is at least twenty-one in Colorado, I would hold that Kilo's sniff of the odors in the free air outside McKnight's car did not violate the Federal or State Constitutions.

II. Analysis

¶92 At the outset, I note that there is no dispute that McKnight's car was lawfully stopped. The parties also agree that Kilo's sniff did not unreasonably prolong the stop, *422that Kilo never set paw in McKnight's car, and that Kilo and his handler remained at all times in a public place where they had a lawful right to be. Additionally, it is uncontested that marijuana continues to be contraband under federal law. Given these circumstances, my colleagues in the majority do not appear to contest that Kilo's sniff is not a search under the Fourth Amendment. Id. at ¶ 37.

¶93 The majority nevertheless throws out McKnight's convictions for possession of methamphetamine and paraphernalia because one of the five drugs Kilo was trained to alert to was marijuana, and Amendment 64 legalized the possession of up to an ounce of marijuana by adults who are at least twenty-one. To be clear, McKnight had no marijuana in his car. But because he could have had marijuana in his car, and because it could have been in an amount and under the limited circumstances permitted by state law, the majority concludes that Kilo's sniff intruded upon a reasonable expectation of privacy. Though I fully embrace the Chief Justice's dissent, I do not take issue with this determination for purposes of my dissent. I focus, instead, on what I view as the most glaring flaw in the majority opinion-the lack of meaningful analysis about whether McKnight had a reasonable expectation of privacy in the odors that emanated from noncontraband items or lawful activity in his car and exited into the boundless airspace. After concluding that McKnight did not have a reasonable expectation of privacy in the infinite public air outside his car, I discuss the strained inference the majority draws from Caballes . I then refute the majority's inapt comparison of this case to Kyllo . And I end by demonstrating that, no matter the majority's insistence to the contrary, its decision is rooted in Fourth Amendment jurisprudence-prominently featuring three U.S. Supreme Court cases, Katz , Caballes , and Kyllo -and nothing in the Colorado Constitution can insulate it from attack.

A. There is No Reasonable Expectation of Privacy in an Odor That Escapes from a Car

¶94 Does a driver have a reasonable expectation of privacy (under the Colorado Constitution or any other authority) in an odor emitted by a noncontraband item or lawful activity in his car after it has escaped into the public airspace outside? The majority sidesteps the question and holds instead that a sniff from a drug-detection dog trained to alert to marijuana is a search under article II, section 7 because, following the passage of Amendment 64, "that sniff can detect lawful activity, namely the legal possession of up to one ounce of marijuana by adults twenty-one and older." Id. at ¶ 7. Thus, my colleagues in the majority focus exclusively on the reasonable expectation of privacy in the possession of an ounce or less of marijuana by anyone in Colorado who is at least twenty-one.

¶95 But that cannot be the end of the discussion. Doesn't the place sniffed matter? Doesn't it matter whether a dog sniff occurs inside a home, in an airport, in a public park, in a correctional institution, outside a car, or somewhere else? And, as relevant here, doesn't it matter that Kilo never entered McKnight's car but merely sniffed the public air outside? Without meaningfully considering the place sniffed and whether McKnight had a reasonable expectation of privacy there, we cannot properly discern if Kilo's sniff violated the Fourth Amendment or article II, section 7. Yet, without conducting the required analysis, today's decision prohibits all sniffs in Colorado by dogs with Kilo's training, unless there is probable cause to conduct a search.1 I undertake here the analysis the majority ignores.

¶96 The jumping-off point is the test articulated by the U.S. Supreme Court in Katz more than fifty years ago to determine whether a search occurred for purposes of the Fourth Amendment-namely, whether the government's activities violated a reasonable expectation of privacy and thus constituted a search within the meaning of the Fourth Amendment.

*423Katz , 389 U.S. at 361, 88 S.Ct. 507 (Harlan, J., concurring).2 On this, there is no disagreement. See maj. op. ¶ 22. Under Katz , no search occurs for Fourth Amendment purposes unless "the individual manifested a subjective expectation of privacy" that "society [is] willing to recognize ... as reasonable." California v. Ciraolo , 476 U.S. 207, 211, 106 S.Ct. 1809, 90 L.Ed.2d 210 (1986) ; see also Katz , 389 U.S. at 361, 88 S.Ct. 507 (Harlan, J., concurring). In Katz , government agents placed an electronic listening and recording device on the outside of a public telephone booth in order to eavesdrop on Katz's conversations. Katz , 389 U.S. at 348, 88 S.Ct. 507 (majority opinion). The Court held that this was a search because Katz had "justifiably relied" on the privacy of the telephone booth and thus had a reasonable expectation of privacy. Id. at 353, 88 S.Ct. 507.

¶97 Of course, "[w]hat a person knowingly exposes to the public, even in his own home ..., is not a subject of Fourth Amendment protection." Id. at 351, 88 S.Ct. 507 ; see also id. at 361, 88 S.Ct. 507 (Harlan, J., concurring) ("[O]bjects, activities, or statements that [a person] exposes to the 'plain view' of outsiders are not 'protected' because no intention to keep them to himself has been exhibited"; conversations in the open, which may be overheard, are not protected either because "the expectation of privacy under the circumstances would be unreasonable."). But Katz did not expose his telephone conversation to the public. To the contrary, he occupied a telephone booth, shut the door behind him, and paid to place a call. Id. at 361, 88 S.Ct. 507 (Harlan, J., concurring). In other words, his phone conversation was intentionally contained within the privacy of the booth. These "critical fact[s]" related to Katz's conduct "surely entitled" him "to assume that his conversation [was] not being intercepted." Id. (internal quotation mark omitted).

¶98 The same cannot be said for anyone who exposes a smell (including from noncontraband items and lawful activity in his car) to the public. Unlike Katz, who "sought to exclude ... the uninvited ear" when he entered the booth and shut the door behind him, id. at 352, 88 S.Ct. 507 (majority opinion), an individual who exposes an odor-producing substance like marijuana to the public air cannot reasonably have intended to exclude the uninvited nose. Certainly that individual has not manifested an expectation of privacy in the public air that society is willing to accept as reasonable. Thus, the fact that a driver may lawfully possesses marijuana in his car pursuant to Colorado law does not mean that he must be deemed to have a reasonable expectation of privacy in the odors that emanate from it and escape into the infinite outside air.

¶99 Two companion cases decided by the U.S. Supreme Court in 1986 are instructive. In the first, Ciraolo , the Court held that naked-eye aerial observations of a backyard from a low-flying airplane do not constitute a search under the Fourth Amendment. 476 U.S. at 215, 106 S.Ct. 1809. The Ciraolo Court emphasized that the challenged observations had taken place "within public navigable airspace" and "in a physically nonintrusive manner," as officers trained in marijuana identification flew over Ciraolo's house in a private plane to observe with their naked eyes what the tall fences he had installed around his backyard blocked from their view. Id. at 213, 106 S.Ct. 1809. Similarly, here, Kilo's sniff took place in the public airspace and in a physically nonintrusive manner. But, importantly, McKnight was in his car, which, as I discuss in more detail later, is significantly less protected by the Fourth Amendment than Ciraolo's home. And there is no evidence that McKnight took any steps exhibiting a subjective *424expectation of privacy that can be compared to the tall fences installed by Ciraolo to block the public's view of his backyard. True, the case before us does not involve naked-eye observations or, more appropriately, naked-nose ones; rather, it entails an investigative tool that enhances a human sense. But as Dow Chemical Co. v. United States (the second case) illustrates, the U.S. Supreme Court has found that the government's use of certain minimally-intrusive, sense-enhancing techniques does not amount to a search under the Fourth Amendment.

¶100 In Dow Chemical , the Court held that the use of a sophisticated precision mapping camera to obtain enhanced aerial photographs of an industrial complex from lawfully navigable airspace is not a search prohibited by the Fourth Amendment. 476 U.S. 227, 229, 239, 106 S.Ct. 1819, 90 L.Ed.2d 226 (1986).3 The Court acknowledged that the warrantless "surveillance of private property by using highly sophisticated surveillance equipment not generally available to the public, such as satellite technology," may well be unconstitutional. Id. at 238, 106 S.Ct. 1819. But the Court determined that "[t]he mere fact that human vision [was] enhanced somewhat, at least to the degree" involved there, did not "give rise to constitutional problems" because the enhanced photographs, although providing the government much more detailed information than naked-eye views, were "not so revealing of intimate details as to raise constitutional concerns." Id. The Court distinguished the challenged activity from the use of electronic equipment that would penetrate walls or windows and allow the government to hear and record confidential discussions inside the plant regarding chemical formulae or other trade secrets; the latter "would raise very different and far more serious questions." Id. at 239, 106 S.Ct. 1819.4

¶101 Just as the magnifying aerial camera in the plane in Dow Chemical allowed officers to discern objects that they could not observe with their naked eyes, Kilo's sniff allowed officers to discern odors in the public air outside McKnight's car that they could not smell with their naked noses. Further, like the technique employed in Dow Chemical , Kilo's sniff took place in lawful airspace and did nothing more than "enhance[ ] somewhat" a human sense-smell instead of sight-without revealing such "intimate details" as to raise constitutional concerns, see id. at 238, 106 S.Ct. 1819. Kilo's alert simply conveyed to his handler that he smelled the odor of at least one of five drugs, including potentially marijuana, in the public air surrounding McKnight's car. This is not the *425type of intimate detail associated with family privacy and the home that historically has raised constitutional concerns.5

¶102 That in some situations only a sniff by a trained narcotics dog could detect the odor of marijuana outside a car is of no moment. Other courts agree. As Judge Lewis T. Babcock explained in United States v. Broadway , "The fact that the smell ... [is] detected by a dog rather than a human does not change its fundamental non-private nature." 580 F. Supp. 2d 1179, 1191 (D. Colo. 2008) ; see also United States v. Roby , 122 F.3d 1120, 1124-25 (8th Cir. 1997) ("The fact that the dog, as odor detector, is more skilled than a human does not render the dog's sniff illegal."); United States v. Goldstein , 635 F.2d 356, 361 (5th Cir. 1981) ("The agents' use of a canine's more enhanced (through training) olfactory sense cannot convert a sniff of the exterior of ... suitcases into a search."). Once the odor of marijuana becomes intermingled with the public airspace, it cannot be deemed private, and anyone who exposes such odor to the public airspace cannot claim to have a reasonable expectation of privacy in it. "Just as evidence in the plain view of officers may be searched without a warrant, evidence in the plain smell may be detected without a warrant." Roby , 122 F.3d at 1125 (internal citation omitted).

¶103 I am concerned that my colleagues in the majority bypass the critical issue related to the location of the sniff, choosing instead to turn a blind eye to the elephant in the room. Equally concerning, the authority on which they lean cannot hold up to scrutiny. Other than the unjustified inference from Caballes and the ill-suited comparison to Kyllo , both of which I address later, the majority relies on: (1) a statement we made thirty-three years ago in People v. Unruh , 713 P.2d 370, 376 (Colo. 1986), abrogated by Esparza , ¶ 11, 272 P.3d at 370, which actually confirms my point : "[A] small minority of courts[ ] have characterized a dog sniff as a search";6 (2) the concurring opinion in a case from the United States Court of Appeals for the Second Circuit from forty-four years ago, which disagreed with the conclusion that case's majority reached-"that use of a marijuana-sniffing dog to ascertain the contents of a private bag amounts to ... 'plain smell,' " which is "comparable to a 'plain view' " and is not a search, see United States v. Bronstein , 521 F.2d 459, 464 (2d Cir. 1975) (Mansfield, J., concurring); and (3) a passing observation by Justice Souter in a footnote in his dissenting opinion in Caballes . Respectfully, this brittle foundation cannot bear the weight the majority stacks on top of it.

¶104 Significantly, I am by no means alone in concluding that there is no reasonable expectation of privacy in odors in the public airspace that emanate from inside a lawfully stopped car and are detected by a narcotics dog. To the contrary, I am in the company of the great bulk of courts that have addressed this issue. See, e.g., Hearn v. Bd. of Pub. Ed. , 191 F.3d 1329, 1332 (11th Cir. 1999) ("[T]he [U.S.] Constitution does not provide Hearn with any expectation of privacy in the odors emanating from her car."); United States v. Morales-Zamora , 914 F.2d 200, 205 (10th Cir. 1990) ("[W]e find that when the odor of *426narcotics escapes from the interior of a vehicle, society does not recognize a reasonable privacy interest in the public airspace containing the incriminating odor."); United States v. Corrujedo , No: 5:17-CR-24-RWS-CMC, 2017 WL 8809641, at *18 (E.D. Tex. Dec. 22, 2017) ("Defendants have no privacy interest in the free air around the car when it is detained incident to a traffic stop" and a dog sniff is conducted.); United States v. Holleman , No. CR12-0040, 2012 WL 3779077, at *15 (N.D. Iowa Aug. 30, 2012) ("Defendant had no reasonable expectation of privacy" in the hotel parking lot where a dog sniff was performed on the exterior of his car.); United States v. Cruz , 1:07-cr-145-WSD, 2008 WL 11383985, at *2 (N.D. Ga. May 15, 2008) ("[T]he Fourth Amendment does not protect what a person knowingly exposes to the public. This 'plain view' exception to the Fourth Amendment's general warrant requirement extends to odors emanating from a defendant's vehicle, giving law enforcement officers probable cause to search the vehicle.") (internal citation omitted); United States v. Taylor , 955 F. Supp. 763, 768-69 (E.D. Mich. 1997) (Because the dog sniff was not conducted on any part of the body of any passenger and no entry was made into the interior of the vehicle, "the use of this dog to sniff the air adjacent to the perimeter of this vehicle, as it stood lawfully-and briefly-detained ..., did not constitute a search within the meaning of the Fourth Amendment."); Merrett v. Dempsey , No. TCA 84-7198-WS, 1992 WL 700220, at *8 (N.D. Fla. May 18, 1992) ("Whether vehicles were occupied or not, whether odors emanated from persons [inside] or not, the motorists ... had no expectation of privacy in the airspace surrounding their vehicles."); State v. Bergmann , 633 N.W.2d 328, 334-35 (Iowa 2001) ("[W]e are persuaded by the ... long-standing viewpoint" that a dog sniff of the perimeter of a lawfully-stopped vehicle in a public place does not constitute a search because "the airspace around the car is not an area protected by the Fourth Amendment.") (alteration and quotation omitted); Jackson v. State , 190 Md.App. 497, 988 A.2d 1154, 1159 (2010) ("The dog is as free to smell cocaine or marijuana as the officer is free to smell the roses or the garbage or 'the breath of new mown hay,' " for "[n]either dog nor man needs a judicial permission slip to sniff the air."); State v. Hartzell , 156 Wash.App. 918, 237 P.3d 928, 934 (2010) (agreeing with the trial court that "Hartzell did not have a reasonable expectation of privacy in the air coming from the open window of the vehicle," as he was not in the car "when the dog sniffed from a lawful vantage point outside the vehicle").

¶105 The majority admits that these cases "have come to a different conclusion" than the one it reaches today concerning Kilo's sniff. Maj. op. ¶ 47. However, it asserts that they are inapposite because none of them involved "legalized marijuana." Id. This is a red herring. While those jurisdictions had not passed a law like Amendment 64, none of the decisions turned on that aspect of the law. The cases I cite in the previous paragraph had absolutely nothing to do with the legalization or decriminalization of marijuana or with the contraband/non-contraband distinction. Rather, they all concluded that a person lacks a reasonable expectation of privacy in odors that escape from his car into the public air outside and are detected by a narcotics dog. The majority's point would have been valid if the cases I cite had been resolved based exclusively on the narcotics dog's ability to detect only the presence or absence of contraband.

¶106 The majority likewise observes that in one of the ten cases I cite, Cruz , "the officers already had 'considerable information' that an individual was engaged in illegal narcotics activity," and in another, Corrujedo , "the defendant consented to a hand search of the vehicle that began before a dog was even deployed." Id. at ¶ 47 n.6. Be that as it may, these circumstances in no way affected the determinations that are relevant for our purposes. In Cruz , the court ruled that the "plain view" exception to the warrant requirement includes "odors [that] emanat[e] from a defendant's vehicle," 2008 WL 11383985, at *2, and in Corrujedo , the court found that there is "no privacy interest in the free air" outside a lawfully stopped vehicle, 2017 WL 8809641, at *18. In any event, Cruz and Corrujedo account for only two of the ten cases I cite. And my list of cases is *427certainly not exhaustive. In fact, as best I can tell, the majority's decision is in conflict with every case during the last three-plus decades that has examined whether a driver has a reasonable expectation of privacy in the odors sniffed by a narcotics dog in the public air outside his lawfully stopped car.7

¶107 Perhaps more importantly, the majority fails to cite a single state or federal case that has gone the other way and has held that a driver has a reasonable expectation of privacy in the odors that escape from his car and become part of the public airspace sniffed by a narcotics dog. This speaks volumes about the outlier nature of the majority's decision.

¶108 Courts also routinely conclude that dog sniffs of odors emanating from other spaces do not constitute searches. The majority goes out of its way to note that these cases are distinguishable. Maj. op. ¶ 47, ¶ 47 n.6. These cases are distinguishable, of course; I cite them here simply to show that, even in other factual contexts, courts have been extremely reluctant to conclude that there is a reasonable expectation of privacy in odors floating in the public air. See , e.g. , United States v. Gant , 112 F.3d 239, 241 (6th Cir. 1997) ("A passenger on a common carrier has no reasonable expectation of privacy in the exterior of his luggage or the airspace surrounding it."); Roby , 122 F.3d at 1125 ("Mr. Roby had an expectation of privacy in his Hampton Inn hotel room. But because the corridor outside that room is traversed by many people, his reasonable privacy expectation does not extend so far. Neither those who stroll the corridor nor a sniff dog needs a warrant for such a trip."); United States v. Duarte , No. 94-1234, 1995 WL 296327, at *3 (6th Cir. May 15, 1995) (A dog sniff does not require probable cause because "there is no reasonable expectation of privacy in the airspace surrounding an object."); United States v. Garcia , 42 F.3d 604, 606 (10th Cir. 1994) (upholding a dog sniff of Garcia's luggage in the baggage car of an Amtrak train because Garcia's reasonable expectation that the contents of his luggage would not be exposed in the absence of consent or a search warrant "did not extend to the air surrounding the luggage"); United States v. Harvey , 961 F.2d 1361, 1363 (8th Cir. 1992) (The defendants had "no reasonable expectation of privacy in the ambient air surrounding their luggage," which was "all that Jupp invaded."); Goldstein , 635 F.2d at 361 ("[T]he passenger's reasonable expectation of privacy" in the contents of her luggage "does not extend to the airspace surrounding that luggage."); United States v. Lucas , 338 F. Supp. 3d 139, 165 (W.D.N.Y. 2018) (joining other circuits in concluding that "no expectation of privacy to the common area outside the [storage] locker would be reasonable"); Khachatourian v. Hacienda La Puente Unified School Dist., No. CV 10-08436 SJO (RZx), 2012 WL 12877986, at *9 (C.D. Cal. Jan. 24, 2012) ("Even if he had a reasonable expectation of privacy in the contents of the counter drawers, Plaintiff had no expectation of privacy in the smell of the air around the counter."); Broadway , 580 F. Supp. 2d at 1191 ("A dog ... does not detect anything inside a home, but merely detects the particulate odors that have escaped from a home," and such odors "are no longer 'private,' but instead are intermingled with 'the public airspace containing the incriminating odor.' ") (quoting Morales-Zamora , 914 F.2d at 205 ).

¶109 Rather than join the great weight of authority, the majority charts an unfamiliar course by drawing an unwarranted inference from the U.S. Supreme Court's decision in Caballes . I turn my attention to this questionable aspect of the majority opinion next.

B. The Inference the Majority Draws from Illinois v. Caballes is Unwarranted

¶110 To place Caballes in context, I must first discuss its predecessor, United States v. Place , 462 U.S. 696, 103 S.Ct. 2637, 77 L.Ed.2d 110 (1983), another Fourth Amendment case from the U.S. Supreme Court discussed by the majority. In Place , after recognizing that a person has a privacy interest in the contents of personal luggage, the Court held that a dog sniff of the outside of *428the luggage did not violate the Fourth Amendment. 462 U.S. at 707, 103 S.Ct. 2637. The Court explained that a dog sniff is "sui generis ," or unique, for two reasons: (1) "the manner in which information is obtained" through a dog sniff "is much less intrusive than a typical search," as it "does not require opening the luggage" and it "ensures that the owner of the property is not subjected to the embarrassment and inconvenience entailed in less discriminate and more intrusive investigative methods";8 and (2) a dog sniff "discloses only the presence or absence of narcotics, a contraband item," and there is no legitimate expectation of privacy in contraband items. Id.

¶111 Close to twenty years later, when confronted with a dog sniff of the exterior of a car in Caballes , the Court affirmed Place -but only found it necessary to rely on the second part of its rationale: "[A]ny interest in possessing contraband cannot be deemed 'legitimate' " and thus a dog sniff that only detects the presence or absence of contraband is not a search under the Fourth Amendment. 543 U.S. at 408, 125 S.Ct. 834. If this holding by the U.S. Supreme Court sounds familiar, it should: it serves as the blueprint for the majority's decision.

¶112 Importantly, the Caballes Court did not disavow the first part of Place 's rationale. The other traits of a dog sniff-that it is limited in scope, minimally intrusive, very fast, and unlikely to cause embarrassment or inconvenience-remain true after Caballes (especially when the sniff involves the public air outside a car). As well, the Court's view of a dog sniff as sui generis , predicated in part on those traits, has not changed. Yet, the majority infers the opposite. It presupposes that, but for the second rationale set forth in Place , the Caballes Court would have concluded that a dog sniff of the exterior of a car constitutes a search. In other words, according to the majority, since the Court held in Caballes that the dog sniff was not a search because it could only detect contraband, the Court must have also meant that a dog sniff like the one here, which could detect a drug the majority now views as noncontraband, must necessarily be a search. See maj. op. ¶ 27. But the reality is that the Court did not have to address the question we confront today because the dog in Caballes could only detect contraband. I am not comfortable making the leap the majority does. Instead, I would apply the Katz test to determine whether McKnight had a reasonable expectation of privacy in the odors floating in the air sniffed by Kilo.

¶113 Not only is the majority's reading of Caballes inconsistent with Katz , it also risks undesirable, if not absurd, results because there is no way to place parameters around an expectation of privacy in the limitless public airspace. What if Kilo had been trained to detect the odor of narcotics while walking a few feet away from McKnight's car? Would that still be a search requiring probable cause? What if the distance had been five or ten feet? What if Kilo had remained in the patrol car during the sniff? Does the majority believe that Amendment 64 gave McKnight a legitimate expectation of privacy in all of the infinite air outside his car? Rather than answer these questions or at least address whether McKnight had a reasonable expectation of privacy in the odors that escaped from his car into the endless public air outside, the majority analogizes Kilo's sniff to the thermal imager used in Kyllo . Id. at ¶ 44 (citing Kyllo , 533 U.S. at 40, 121 S.Ct. 2038 ). But, as I demonstrate next, Kyllo cannot carry the day for the majority.

C. The Majority Erroneously Analogizes this Case to Kyllo

¶114 The majority's suspect analogy to Kyllo falls flat. A dog sniff is nothing like a *429thermal-imaging scanner that detects infrared radiation and converts it into images based on relative warmth. More importantly, there is a world of difference in the Fourth Amendment arena between obtaining information about the interior of Kyllo's home, including potentially "at what hour each night the lady of the house takes her daily sauna and bath," Kyllo , 533 U.S. at 38, 121 S.Ct. 2038, and obtaining information about any odors emitted by the noncontraband contents and lawful activities in McKnight's car after they have escaped into the public air outside during a lawful traffic stop.

1. A Dog Sniff Can Never Be Confused With a Thermal Imager

¶115 As relevant here, the majority stumbles out of the gate when it likens Kilo to a thermal-imaging scanner.9 Maj. op. ¶¶ 44-46. Kilo is not a robot or the product of a laboratory experiment. Nor does Kilo function on technology. Kilo is a dog-a living, walking, sniffing animal. That Kilo has been trained to detect the smell of five drugs with his nose does not make him technology-like.

¶116 Certainly "a drug sniffing dog is not 'technology' of the type addressed in Kyllo ." Bergmann , 633 N.W.2d at 334 ; see also Duarte , 1995 WL 296327, at *3 ("[T]he use of a dog to detect the odor of contraband is not analogous to the use of binoculars or magnifying devices to observe objects or activities not exposed to public view."). Indeed "a dog's sense of smell, while more acute than a human's, does not compare to a technology that can turn minute gradations in temperature into video tapes from 1500 feet away." United States v. Ishmael , 843 F. Supp. 205, 213 (E.D. Tex. 1994), rev'd on other grounds , 48 F.3d 850 (5th Cir. 1995). " Kyllo specifically speaks to highly advanced technology that is not readily available to the general public," which is used to gain access to details inside a home that would be "unknowable without physical intrusion." Bergmann , 633 N.W.2d at 334 (citation omitted). Contrary to what the majority implies, "It cannot seriously be argued that a dog can explore 'details of the home' that would 'previously have been unknowable' without physical intrusion." Broadway , 580 F. Supp. 2d at 1191 (quoting Kyllo , 533 U.S. at 40, 121 S.Ct. 2038 ). "To the extent a dog can detect a scent, ... it does not detect anything that 'would have been unknowable' without physical intrusion when the Fourth Amendment was adopted in 1791." Id. (quoting Kyllo , 533 U.S. at 40, 121 S.Ct. 2038 ).

2. For Fourth Amendment Purposes, There Is No Place Like Home

¶117 The majority acknowledges that the holding in Kyllo hinged in part on the fact that the challenged search was of a home, not a car. Maj. op. ¶ 45. Yet, in the next breath, it seemingly robs that difference of any meaning and adopts a one-size-fits-all expectation-of-privacy standard for homes and cars by relying on Kyllo just the same. See id. If, as the majority admits, the Fourth Amendment offers a home heightened protection, Kyllo 's rationale cannot apply with equal force in the context of a car. Unlike the majority, I readily recognize that this case cannot be shoehorned under the Kyllo umbrella. Consequently, I attempt to do justice to the crystal-clear distinction the U.S. Supreme Court has consistently drawn in the Fourth Amendment framework between the home and other places.

¶118 The right of a person "to retreat into his own home and there be free from unreasonable governmental intrusion" is at the heart of the Fourth Amendment. Silverman v. United States , 365 U.S. 505, 511, 81 S.Ct. 679, 5 L.Ed.2d 734 (1961). "[W]hen it comes to the Fourth Amendment, the home is first among equals." Jardines , 569 U.S. at 6, 133 S.Ct. 1409. The greater expectation of privacy in the interior of a home is moored in "the overriding respect for the sanctity of the home that has been embedded in our traditions since the origins of the Republic." Payton v. New York , 445 U.S. 573, 601, 100 S.Ct. 1371, 63 L.Ed.2d 639 (1980). Indeed, physical intrusion into the home "is the chief evil *430against which the ... Fourth Amendment is directed." United States v. U.S. Dist. Ct. , 407 U.S. 297, 313, 92 S.Ct. 2125, 32 L.Ed.2d 752 (1972). Because the home is the most sacred of Fourth Amendment spaces, with few exceptions, a warrantless search of a home is unreasonable and unconstitutional. See Jardines , 569 U.S. at 6, 133 S.Ct. 1409 ; Kentucky v. King , 563 U.S. 452, 474, 131 S.Ct. 1849, 179 L.Ed.2d 865 (2011) (Ginsburg, J., dissenting) ("In no quarter does the Fourth Amendment apply with greater force than in our homes, our most private space which, for centuries, has been regarded as 'entitled to special protection.' ") (quoting Georgia v. Randolph , 547 U.S. 103, 115, 126 S.Ct. 1515, 164 L.Ed.2d 208 (2006) ).

¶119 In Kyllo , the Court reiterated the unique protection afforded the interior of a home by the Fourth Amendment. It was in part on this basis that the Court rejected the prosecution's argument "that the thermal imaging was constitutional because it did not 'detect private activities' " in any private areas:

The Fourth Amendment's protection of the home has never been tied to measurement of the quality or quantity of information obtained. In Silverman, for example, we made clear that any physical invasion of the structure of the home, "by even a fraction of an inch," was too much , and there is certainly no exception to the warrant requirement for the officer who barely cracks open the front door and sees nothing but the nonintimate rug on the vestibule floor. In the home, our cases show, all details are intimate details, because the entire area is held safe from prying government eyes.

Kyllo , 533 U.S. at 37, 121 S.Ct. 2038 (emphases added except as to the word "all") (internal citation omitted). Inasmuch as the temperature inside Kyllo's residence was a detail of the home, it was an intimate one as to which he had a reasonable expectation of privacy. Id. at 38, 121 S.Ct. 2038. So long as the detail learned through the highly-advanced technological device used was one "that would previously have been unknowable without physical intrusion, the surveillance [was] a 'search' and [was] presumptively unreasonable without a warrant." Id. at 40, 121 S.Ct. 2038.

¶120 Conversely, the U.S. Supreme Court has recognized that "the expectation of privacy with respect to one's automobile is significantly less than that relating to one's home or office." South Dakota v. Opperman , 428 U.S. 364, 367, 96 S.Ct. 3092, 49 L.Ed.2d 1000 (1976) (emphasis added); see also Morales-Zamora , 914 F.2d at 205 (indicating in a dog sniff case that "there is a lesser expectation of privacy in a vehicle than in a home"). In fact, the Court has acknowledged that "since the beginning of the government" there has been "a necessary difference between a search of a ... house ... and a search of a ship, motor boat, wagon, or automobile." Carroll v. United States , 267 U.S. 132, 153, 45 S.Ct. 280, 69 L.Ed. 543 (1925). Our court has likewise observed that there is a diminished expectation of privacy in automobiles. People v. Litchfield , 918 P.2d 1099, 1103 (Colo. 1996).

¶121 Because the majority fails to give effect to the substantial differences between this case and Kyllo , its analogy of the dog sniff here to the thermal imager there misses the mark. Perhaps concerned with the vulnerability of this comparison and the unreasonable inference it draws from Caballes , the majority attempts to use our State Constitution as a bulwark for its decision. But, as I address next, its efforts are for naught.

D. The Majority's Decision is Clearly Grounded in Fourth Amendment Jurisprudence

¶122 Despite relying almost exclusively on U.S. Supreme Court precedent applying the Fourth Amendment for its analysis, the majority attempts to shelter its decision in the Colorado Constitution: "[W]e rest our holding on the Colorado Constitution, not on the Fourth Amendment." Maj. op. ¶ 47. At the end of its opinion, the majority doubles down on this proclamation and goes so far as to take the unusual step of issuing a disclaimer: "This holding is based solely on the protections afforded by our state constitution, and to the extent we discuss any federal cases in coming to our conclusion, it is only for the purpose of analogy and comparison."

*431Id. at ¶ 63. Indeed, by relying on its Colorado-only justification, the majority dismisses the lion's share of federal and state cases that have concluded there is no reasonable expectation of privacy in odors in the public air outside a car that are sniffed by a narcotics dog, even if they may have originated inside the car. See id. at ¶ 47. The majority's own opinion undercuts its Colorado-only claim.

¶123 Respectfully, because the majority hitched its wagon to Fourth Amendment jurisprudence, it cannot seek to barricade its opinion behind the mantle of the Colorado Constitution. That the majority finds it necessary to attempt to convince us that its discussion of any federal authority (the authority on which it principally leans) is strictly "for the purpose of analogy and comparison" is telling. Id. at ¶ 63.

¶124 The majority appears to conflate two different concepts: (1) the creation through a state constitutional amendment of a privacy interest in the possession of marijuana under certain limited circumstances; and (2) the analytical framework to determine what constitutes a search when that privacy interest is involved. I try to unravel any confusion that may exist.

¶125 In the past this court has interpreted the Colorado Constitution to protect certain privacy interests beyond those protected by the Fourth Amendment. See People v. Corr , 682 P.2d 20, 27-28 (Colo. 1984) (telephone toll records); People v. Sporleder , 666 P.2d 135, 140-41 (Colo. 1983) (outgoing calls monitored by pen registers); Charnes v. DiGiacomo , 200 Colo. 94, 612 P.2d 1117, 1119-21 (1980) (bank transactions recorded by bank personnel). For purposes of this dissent, I am not disputing that Amendment 64 created a new privacy interest in the possession of marijuana under limited circumstances in Colorado and that this privacy interest does not exist under federal law. See maj. op. ¶ 42. In fact, for purposes of this dissent, I do not even take issue with the majority's curious conclusion that "Amendment 64 expanded the protections of article II, section 7" to provide this new reasonable expectation of privacy "in the lawful activity of possessing marijuana in Colorado." Id.

¶126 The issue I address in this dissent is whether, given the majority's finding that McKnight had a reasonable expectation of privacy under the Colorado Constitution in the possession of up to an ounce of marijuana , officers violated his state constitutional right to be free from unreasonable searches and seizures by having Kilo sniff the air outside his car. Resolution of this question requires us to determine whether Kilo's sniff is a search-and a search is defined the same way under article II, section 7 as it is under the Fourth Amendment. It is revealing that in addressing the question, the majority essentially relies on three cases, Katz , Caballes , and Kyllo , all of which are U.S. Supreme Court cases interpreting and applying the Fourth Amendment. And, to the extent the majority also relies on Colorado precedent in Esparza , it does so only because we adopted the analytical framework and rationale in Caballes there. Indeed, as the majority concedes, " Esparza ... represented a kind of return [for this court] to the federal fold."10 Id. at ¶ 34.

*432¶127 The majority cites Michigan v. Long , 463 U.S. 1032, 103 S.Ct. 3469, 77 L.Ed.2d 1201 (1983), to justify its disclaimer. But that case is inapposite. While today's opinion certainly "appears to rest primarily on federal law, or to be interwoven with the federal law," there is not "an adequate and independent state ground" to support it. Id. at 1040, 103 S.Ct. 3469. Despite its self-serving characterization of today's decision as embedded exclusively in Colorado law, the majority can point to no "bona fide separate, adequate, and independent [state] grounds" underpinning it. Id. at 1041, 103 S.Ct. 3469. The majority's conclusion that the possession of up to one ounce of marijuana by adults who are at least twenty-one is now protected in Colorado by article II, section 7 cannot do the trick because this provision does not (and cannot) resolve whether Kilo's sniff was a search . There are simply no grounds to allow the majority to tuck its opinion in the protective fold of our State Constitution.

¶128 In short, the majority predominantly relies on Fourth Amendment U.S. Supreme Court jurisprudence in applying Colorado's search-and-seizure constitutional provision. As such, its decision is tethered to Fourth Amendment U.S. Supreme Court jurisprudence, not search-and-seizure case law unique to Colorado. The majority cannot undo this reality with a disclaimer.

III. Conclusion

¶129 I recognize that the United States Supreme Court has not addressed whether, following the passage of a state law like Amendment 64, a driver has a reasonable expectation of privacy in odors that emanate from inside his car and are detected in the public air outside by a narcotics dog with Kilo's training. But considering the advent of the decriminalization and legalization of marijuana in a growing number of states, see maj. op. ¶ 47, I believe that the issue has taken on a new level of relevance. The approach in Caballes no longer suffices. Despite Caballes , state laws decriminalizing or legalizing marijuana can be interpreted so as to render all sniffs by dogs with Kilo's training categorically unlawful. The majority opinion illustrates the point. To my mind, the marijuana sea change requires a recalibration of the balance struck in Caballes between the needs of law enforcement and the interests protected by the Fourth Amendment. This is certainly not a Colorado-specific issue because the majority's search-and-seizure analysis takes root in Fourth Amendment jurisprudence-and specifically in U.S. Supreme Court precedent.

¶130 In the end, I worry that today's decision unnecessarily upsets the applecart. For the reasons articulated here and in the Chief Justice's compelling dissent, I believe that the court of appeals erred. Therefore, I would reverse its judgment. Accordingly, I respectfully dissent.

I am authorized to state that JUSTICE BOATRIGHT joins in this dissent.