United States v. Diaz

Full title: UNITED STATES OF AMERICA, v. ERICK DIAZ, ALEXANDER GEORGE BOZZETTI…

Court: UNITED STATES DISTRICT COURT FOR THE DISTRICT OF SOUTH CAROLINA CHARLESTON DIVISION

Date published: Apr 6, 2018

No. 2:16-cr-00055-DCN (D.S.C. Apr. 6, 2018)

Opinion

No. 2:16-cr-00055-DCN

04-06-2018

UNITED STATES OF AMERICA, v. ERICK DIAZ, ALEXANDER GEORGE BOZZETTI, Defendants.

DAVID C. NORTON UNITED STATES DISTRICT JUDGE

ORDER

This matter is before the court on a motion to suppress filed by defendants Erik Diaz ("Diaz") and Alexander George Bozzetti ("Bozzetti") (collectively, "defendants"). After consideration of the testimony presented during multiple lengthy suppression hearings, a thorough analysis of the relevant caselaw, and a careful review of the dashcam video of the traffic stop, the court grants the motion to suppress. The court finds that there was no reasonable suspicion to prolong the traffic stop to conduct the free-air dog sniff as required under Rodriguez v. United States, 135 S. Ct. 1609 (2015). Furthermore, even if the officers did have reasonable suspicion to prolong the traffic stop, the drug dog Rao ("Rao") did not "alert." And finally, even if Rao did "alert," the court finds that Rao was unreliable under Florida v. Harris, 568 U.S. 237 (2013) such that its "alert" did not establish probable cause to search the vehicle.

I. BACKGROUND

On June 4, 2015, Lance Corporal K.L. Byrd ("Byrd") of the South Carolina Highway Patrol ("SCHP") visually observed a black Nissan with Florida license plates traveling at 77 miles per hour in a 70 miles per hour zone on Interstate 95 and swerving multiple times across the white fog line. Byrd decided to pull the car over for the traffic infraction. See S.C. Code §§ 56-5-1810, 1900. Byrd stopped the car travelling southbound on I-95 at mile marker 68, roughly 9-10 hours from Pennsylvania. Tr. 21:1-16. Equipped with a dashboard video camera, Byrd's vehicle recorded the ensuing traffic stop.

Once the car stopped, Byrd approached the passenger side of the vehicle to find Diaz in the driver's seat, and Bozzetti and Juan Rodriguez ("Rodriguez") as passengers. Byrd asked Diaz for his license and registration and to step outside the vehicle. At this point, Byrd observed an "overwhelming" odor of air freshener from the five air fresheners visible, three clip-in air fresheners in the front air conditioning vents, one on the rear view mirror, and one hanging in the back passenger area. Byrd also noticed that the passenger in the front seat was breathing rapidly. Byrd explained to Diaz the reason for the traffic stop, and began to ask a series of questions about where the car was coming from, to which Diaz responded that the car was coming from Pennsylvania from his brother's wedding. Byrd appeared to have difficulties understanding Diaz, as evidenced by Byrd asking where the car was coming from and Diaz repeating "Pennsylvania" multiple times before Byrd seems to understand the answer.

Byrd then called for backup from his superior in the SCHP "Aggressive Criminal Enforcement" ("ACE") team, Corporal Avery Stephen English ("English"). While Diaz was waiting outside his car, Byrd then went back to his police car to call Lance Corporal James Franklin Sweatman III ("Sweatman") of the SCHP to come to the scene and bring his police dog Rao. After calling Sweatman, Byrd left the police car and informed Diaz that the stop would take longer because he was checking the vehicle identification number ("VIN"). At this point, English had arrived. English then went to talk to the remaining two passengers in the car, Bozzetti and Rodriguez, and asked them questions including: "How long have you been on the road," "How long were you all up in Pennsylvania," and "Are you all related? Are you friends?" The passengers answered while Diaz was standing outside the car. English also asked the passengers questions about the wedding in Pennsylvania that Diaz had told Byrd they all attended on Sunday. When asked, Bozzetti told English that they were in Pennsylvania for the wedding of Diaz's brother and that it was on Saturday. Upon running the VIN number back in the police car, Byrd determined that the rental car that Diaz was driving was due back in Miami the day before.¹ Byrd asked Diaz about the rental car being overdue under the rental agreement, to which Diaz responded that he called the rental company that day to extend the rental period for the car.

1 The rental car was rented in Miami on Wednesday for the wedding that was in Pennsylvania on Saturday or Sunday, and the car was stopped late at night the day after the rental car was due to be returned in Miami.

Within fifteen minutes of first making contact with Diaz, Byrd handed Diaz's documents back and explained the warning to him. After giving the warning, Byrd asked Diaz if he could search the vehicle. Diaz refused, and said that he would like to get back on the road to Miami. Byrd then said that Rao would run the exterior of the car and if it did not alert to narcotics, Diaz would be free to go. Diaz responded "Okay." Sweatman then deployed Rao to conduct a free-air sniff of the vehicle to determine if there were narcotics in the vehicle. After being run around the car, Rao "alerted" to the driver's door of the vehicle. The troopers asked the passengers if they were "smoking a little marijuana, to tell us and we will let you be on your way." The passengers all once again denied that there were any drugs in the car. Upon searching the car after Rao "alerted," the officers found no narcotics but did find 171 credit and gift cards contained inside luggage and empty cigarette packs.

Defendants Diaz and Bozzetti were indicted on one count of possession of counterfeit access devices, a violation of Title 18, United States Code § 1029(a)(3) and (2), and one count of aggravated identity theft, a violation of 18 U.S.C. § 1028(a)(1). On January 13, 2017, Bozzetti filed a motion to suppress the evidence found in the car after the traffic stop. On April 12, 2017, Diaz joined the motion. The government responded on May 1, 2017. The court held a hearing on October 27, 2017, during which it heard the testimony of English and Sweatman. Because Byrd could not be present at the first suppression hearing, the court continued the hearing until February 1, 2018, during which Byrd testified. The motion has been fully briefed and is now ripe for the court's review.

II. DISCUSSION

Defendants challenge the seizure of the evidence in the case on three separate grounds: (1) that the initial traffic stop was not supported by reasonable suspicion; (2) there was not reasonable suspicion to warrant the additional detention after the traffic stop had been completed; (3) neither the drug dog nor the alert were reliable enough to establish probable cause. The court analyzes each argument in turn.²

2 The court issued an oral order to sequester the witnesses at the beginning of the first suppression hearing on October 27, 2017. Tr. 3:18-21. This hearing was continued because Byrd was not present at the hearing, and resumed on February 1, 2018. The day before his testimony, Byrd spoke to English about the facts of this case. Tr. 185:15-186:2. Diaz asked for the court to strike the testimony of both Byrd and English entirely, arguing that this communication was a clear violation of the sequestration order that the court imposed at the beginning of the first suppression hearing in October 2017. Tr. 236:1-8. Counsel for the government countered that he "wasn't aware of a sequestration order at the beginning of the case," and if there was a sequestration order that he "assumed that [the law enforcement witnesses] were in court to hear it" but that he "[didn't] know if [he] talked about it otherwise." Tr. 236:18-19.
The court first stresses to the government the importance of communicating the requirements of sequestration to its witnesses—especially to those witnesses such as Byrd who were not physically present at the first suppression hearing to hear the court's oral order. But it would be untenable for the court to strike the testimony of two out of the three law enforcement officers who testified during the suppression hearing as a sanction for Byrd and English's violation of that sequestration order. For one, English had not yet violated the sequestration order at the time that he gave his testimony. And Byrd did not know about the sequestration order at the time that he communicated with English because the government had never informed him of it. Therefore, the court denies Diaz's motion to strike the testimony of Byrd and English.

A. Bozetti's Standing to Challenge Search

The government first argues that Bozetti, as a passenger of the rental car whose name was not on the rental car agreement,³ has no legitimate expectation of privacy in the contraband found under the front passenger side floor mats. Contraband was found in three locations: in the defendants' luggage, in several empty cigarette packets, and under the front passenger side floor mats. It is the contraband found in this last category that the government challenges Bozetti's standing to contest.⁴

3 The parties do not contest the fact that Bozetti is not listed as an authorized driver on the rental car agreement.

4 Certainly, Bozetti as a passenger may challenge the reasonableness of the initial traffic stop. In Brendlin v. California, 551 U.S. 249, 259 (2007), the Court explained that a passenger can challenge the stop of a vehicle, if that stop was unreasonable because "[a] traffic stop necessarily curtails the travel a passenger has chosen just as much as it halts the driver, diverting both from the stream of traffic to the side of the road, and the police activity that normally amounts to intrusion on privacy and personal security does not [ ] distinguish between passenger and driver." Id. at 257.

The government relies on a rule about reasonable expectations of privacy that does not control in this case. Specifically, it relies on circumstances similar to that of United States v. Avagyan, 164 F. Supp. 3d 864, 880 (E.D. Va. 2016), aff'd sub nom. United States v. Ghazaryan, 2017 WL 1382027 (4th Cir. Apr. 18, 2017), where the court held that under Rakas v. Illinois, 439 U.S. 128 (1978), non-owner passengers who have the owner's permission to use the car⁵ have standing to challenge a search only of the "particular areas of the automobile in which he has a legitimate and reasonable expectation of privacy." The Avagyan court went on to say that "[n]on-owner passengers who have the owner's permission to be in the vehicle enjoy standing only as to items over which they have control; passengers, therefore, have no interest in the space under the driver's seat, in the trunk, or in a locked glovebox to which only the owner has a key." The government argues that under the current state of the law Bozetti cannot challenge the search of the front passenger side floor mats because he has no reasonable expectation of privacy in that part of the rental car. Certainly, while the contraband was not found in the trunk, the fact that the cards were found underneath the passenger side floor mats, and that a slit had been cut on the vertical portion of the floorboard below the glove compartment into which the cards appeared to have been hidden, indicates that Bozetti would not have had a legitimate expectation of privacy in that portion of the car and thus has no standing to contest the contraband that was found in that specific location.⁶

5 It is not clear that Bozetti even had Diaz's permission to use the rental car. No testimony was produced at the hearing that Bozetti even drove the car for portions of the trip.

6 The U.S. Supreme Court recently heard oral arguments in the case of Byrd v. United States, on whether drivers who are not listed as authorized drivers on the rental agreement are entitled to the same protections under the Fourth Amendment as those who are included on the agreement. The Court has yet to issue an opinion. Surprisingly, neither party referenced Byrd in the briefing before the court or in their oral arguments. However, Bozetti makes no argument that he was driving the rental car at any point. Certainly, when the traffic stop was initiated Diaz—not Bozetti—was driving the car.

But that is not the correct rule. Here, Bozzetti was a passenger of the car who was seized as soon as Byrd impermissibly extended the traffic stop without reasonable suspicion. As soon as a defendant is illegally seized, that defendant is entitled to suppress any evidence that is derived from the unlawful seizure. In United States v. Rodriguez-Escalera, 2018 WL 1178359, at *7 (7th Cir. Mar. 7, 2018), the government pursued the exact same argument that is currently before the court. There, the government argued that the passenger in a car that was subjected to a free-air dog sniff had no standing to challenge the methamphetamine that was found in the subsequent search of that car once the drug dog alerted, even if the traffic stop had been unconstitutionally prolonged without reasonable suspicion. The Seventh Circuit rejected this argument out of hand, reasoning that because the passenger of the car had been detained without reasonable cause, "[t]he drug evidence was therefore derived from the unlawful seizure, and Rodriguez, as a subject of that seizure, is entitled to have suppressed any evidence which is the fruit of that violation." Id. Here, there is a clear causal connection between Bozzetti's seizure and the officers finding the contraband in the floorboard. Bozzetti was detained in the Nissan while Sweatman ran Rao around the Nissan to conduct a free-air sniff. Rao's alert during this unconstitutionally prolonged seizure of Bozzetti is the only thing that gave the officers probable cause to search the car. Thus, any contraband seized during the subsequent search of the car is inadmissible if it is the fruit of a prolonged unlawful seizure. Bozzetti—like the passenger in Rodriguez-Escalera—has standing to challenge the admission of any evidence which is the fruit of the unlawful seizure.

B. Reasonable Suspicion for Initial Traffic Stop

Having determined that both defendants have standing to challenge the evidence that was seized as a result of the search, the court moves to the merits of the motion. Defendants first challenge the initial traffic stop as unconstitutional, arguing that Byrd lacked reasonable suspicion to stop the car. This argument is unavailing. Byrd "visually observed" Diaz speeding at mile marker 74, but did not turn on his blue lights and initiate the traffic stop until mile marker 68. Tr. 8-15. In those intervening six miles, Byrd also observed Diaz make a lane violation by crossing the white fog line. Tr. 194:19-23. It was only after Byrd observed Diaz cross the white fog line that he stopped the car.

A traffic stop is a "seizure" within the meaning of the Fourth Amendment and must be reasonable under the circumstances. See Delaware v. Prouse, 440 U.S. 648, 653-54 (1979). The constitutionality of a traffic stop is assessed under the two-prong standard in Terry v. Ohio, 392 U.S. 1 (1968). First, the articulated basis for the traffic stop must be legitimate. See United States v. Rusher, 966 F.2d 868, 875 (4th Cir. 1992). Second, the officers' actions during the traffic stop must be "reasonably related in scope" to the basis for the seizure. Id.

This first prong is satisfied "whenever it is lawful for police to detain an automobile and its occupants pending inquiry into a vehicular violation." See Arizona v. Johnson, 555 U.S. 323, 327 (2009). A "vehicular violation" includes a failure to comply with traffic laws. Here, Byrd visually observed the car going 77 miles per hour in a 70 miles per hour speed zone, and observed the vehicle swerve to the right across the solid white fog line. Tr. 140:16-20 (Byrd testifying about pulling the Nissan over for traffic violations, specifically that he "noticed" the Nissan "traveling at a high rate of speed, 70 miles per hour" and then "actually cross the white fog line to the right." Tr. 140:16-20. Both the speeding of the car and the swerving between lanes constitute traffic violations.

In United States v. Sowards, 690 F.3d 583, 591 (4th Cir. 2012), the Fourth Circuit stated in broad terms that "the Fourth Amendment does not allow, and the case law does not support, blanket approval for the proposition that an officer's visual speed estimate, in and of itself, will always suffice as a basis for probable cause to initiate a traffic stop." Where, as here, the traffic stop is based on a vehicle's estimated speed that is only slightly in excess of the legal speed limit, there must be "additional indicia of reliability" such as radar, pacing, or other corroborating evidence to support the reasonableness of the officer's visual estimate. There was no testimony about whether Byrd's so-called "visual inspection" of speeding was corroborated by pacing or radar. Instead, Byrd testified that he determined that the Nissan was speeding by "visually" looking at the vehicle and determining that the Nissan was going faster than 70 miles per hour. Tr. 192:10-12. Byrd did testify that he "did activate" his radar "at some point," although he conceded that there was nothing in his report of the traffic stop that discussed a radar. Tr. 192:13-23. Certainly, Byrd was trained in visually estimating speeds and his testimony regarding Diaz's rate of speed was credible. But based on the "totality of the circumstances" analysis that remains the prevailing standard for evaluating the reasonableness of an officer's visual speed estimate, Byrd's visual speed estimate was in and of itself insufficient to constitute probable cause to initiate a traffic stop.

That being said, Byrd did not initiate the traffic stop until he observed the vehicle travel across the fog line, or the white traffic line along the shoulder of the road. Swerving across the white traffic line is a traffic violation. S.C. Code § 56-5-1810, 1900. During the suppression hearing, no testimony was offered that Diaz did not actually swerve across the fog line. Instead, Diaz pointed to the lack of dashboard video recording of the traffic stop at issue in this case. The dashcam video did not capture the traffic infractions, as Byrd testified that the video activates only when the police cruiser's blue lights turn on. Tr. 140:22-141:22. Furthermore, at the time of the traffic stop SCHP officers did not have any body cameras. Tr. 188:12-14. Despite Diaz's best efforts to argue otherwise, the court has no reason to require that officers use their personal cellular phones to record a routine traffic stop. See Tr. 195:13-19. Since Diaz has neither denied that he committed the traffic violation of swerving across the fog line nor produced any evidence that calls Byrd's testimony into question, the court finds that Byrd had probable cause to initiate the traffic stop of the vehicle.

In Whren v. United States, 517 U.S. 806, 809-810 (1996), the Supreme Court held that a traffic stop did not violate the plaintiff's Fourth Amendment rights where the officer had probable cause to believe several vehicle code violations had been committed. The Whren court stated that "[a]s a general matter, the decision to stop an automobile is reasonable where the police have probable cause to believe that a traffic violation has occurred." Id. Since the car swerved across the white fog line, this offense alone constitutes reasonable suspicion to effectuate a traffic stop. Byrd had reasonable suspicion for the initial traffic stop and the first prong of Terry is fulfilled.⁷

7 Defendants attempt to argue that Byrd's traffic stop was pretextual, and that he did not initiate the traffic stop because of the speeding—which, admittedly, was only seven miles over the speed limit. However, the law is clear that the court may not inquire into an officer's subjective reasons for initiating a stop. Whether Byrd had a pretextual reason for initially stopping the car is irrelevant in determining if there was reasonable suspicion for the stop. United States v. Johnson, 734 F.3d 270, 275 (4th Cir. 2013) (holding that a traffic stop is legitimate "when officers observe a traffic violation, regardless of their true, subjective motives for stopping the vehicle").

The second prong of the Terry test asks if the officer took impermissible actions after initiating the traffic stop. Officers may conduct certain safety-related checks such as requesting a driver's license and vehicle registration, checking for criminal records, and outstanding arrest warrants "so long as" these unrelated inquires did not "measurably extend the duration of the stop." Rodriguez v. United States, 135 S. Ct. 1609, 1615 (2015) (quoting Johnson, 555 U.S. at 333). Here, Byrd asked Diaz for his driver's license and vehicle information, asked Diaz to step out of his car and asked a number of questions for roughly two minutes about his trip. Byrd then checked the VIN, and concurrently asked the passengers of the car the same questions that he had asked Diaz about the trip. Byrd then headed back to his vehicle, and English arrived on the scene. Byrd then "debriefed" English on the situation, shared his concerns based on the passengers' "nervous demeanor" and other factors that they were transporting contraband, after which English asked the passengers the same questions that Byrd had asked. While English was asking questions, Byrd checked for outstanding warrants on the license and began to type out a warning. After printing out the warning, Byrd asked Diaz a number of questions about the presence of contraband in the vehicle as well as questions about his itinerary. Byrd then handed Diaz the warning, and asked if he could search the vehicle. At the point that Byrd handed Diaz the warning, the initial traffic stop was complete. The court now finds that Byrd did not delay the stop to ask questions irrelevant to the purpose of the stop, as Byrd's questions about the purpose and itinerary of the defendants' trip were posed during and were relevant to the legitimate traffic stop. See Muehler v. Mena, 544 U.S. 93 (2005) (Analogizing questioning during a search to performing a dog sniff during a traffic stop, which does not violate the Fourth Amendment if it does not extend the stop "beyond the time reasonably required to complete [the stop's original purpose]."). The Fourth Circuit has cautioned against engaging in "post hoc evaluation of police conduct" to determine if there are less-intrusive means by which the objectives of the traffic stop could have been accomplished. Palmer, 820 F.3d at 649 (quoting United States v. Sharpe, 470 U.S. 675, 686-87 (1985).

Since the traffic stop fulfills both prongs of the Terry test, the initial traffic stop was constitutional and does not constitute a ground for suppression.

C. Reasonable Suspicion to Prolong the Traffic Stop

Next, the defendants contend that the evidence should be suppressed because there was no reasonable suspicion to prolong the traffic stop as required to conduct the free-air dog sniff under Rodriguez.⁸ The court agrees.

8 Diaz had already refused to consent to a search of the car, at which point Byrd stated that Rao would be run around the car. The government makes no argument that the free-air dog sniff was the result of a consent search. Indeed, Byrd admitted that although he asked Diaz for consent to search his car, Diaz was "[n]ot at that point" free to leave as Byrd believed that he had "not enough probable cause at that point" to do a free-air dog sniff. Tr. 168:13-24. Therefore, the dog sniff could not be a search pursuant to consent.

1. Applicable Rule

The government relies on Illinois v. Caballes, 543 U.S. 405 (2005) to argue that the Supreme Court has already held that "law enforcement may use a canine to sniff around a vehicle even in the absence of reasonable, articulable suspicion of drug activity." Gov. Resp. 6. First, this is a misrepresentation of the basic rule articulated in Caballes. In Caballes, the Supreme Court found that the use of a trained narcotics sniffing dog around the exterior of car during an otherwise lawful traffic stop was permissible because it did not implicate privacy interests, and that "the duration of the stop . . . was entirely justified by the traffic offense and the ordinary inquiries incident to such a stop." Furthermore, Caballes is inapplicable here—Byrd had already handed Diaz the warning, thus ending the traffic stop, before the dog sniff was initiated. Caballes says nothing about whether reasonable suspicion is required where a dog sniff is initiated after an "otherwise-completed traffic stop." The applicable rule is the one articulated in Rodriguez. In Rodriguez, the Court addressed the question of "whether the Fourth Amendment tolerates a dog sniff conducted after completion of a traffic stop," and held that "a police stop exceeding the time needed to handle the matter for which the stop was made violates the Constitution's shield against unreasonable seizures." Id. at 1612. The Rodriguez Court discussed the mission of a traffic stop as encompassing the typical "checking of the driver's license, determining whether there are outstanding warrants, and inspecting the automobile's registration and proof of insurance." Id. at 1615.

In Rodriguez, the officer had issued a warning and returned the driver's documents prior to the dog sniff. Similarly, here the dashcam video clearly shows that Byrd had already issued Diaz the warning before the dog sniff was initiated. The warning was issued 15 minutes after Byrd "first made contact" with Diaz. The entire stop took 23 minutes, so the dog sniff extended the stop by 8 minutes. The dog sniff therefore prolonged the stop.⁹ And indeed, English himself testified that Byrd had completed the initial traffic stop before the free-air sniff was conducted, as Byrd had returned all of the passengers' ID cards, registration, and rental car agreement, and explained the warning before Rao was run around the car. Tr. 17:19-18:10. To be constitutionally reasonable, Byrd needed reasonable suspicion to conduct it. As explained below, the court finds that there was no reasonable suspicion.

9 The government argued that Rao had arrived on the scene at the point that Byrd was handing Diaz the warning. However, Rodriguez held that the critical question as to whether a dog sniff during a traffic stop violates the Fourth Amendment is not whether the dog sniff occurs before or after the officer issues the warning or ticket, but rather whether conducting the sniff adds time to the stop. Rodriguez, 135 S.Ct. at 1619-1621. The Rodriguez court specifically held that "a police stop exceeding the time needed to handle the matter for which the stop was made violates the Constitution's shield against unreasonable seizures." Id. at 1612 (emphasis added). Even under the most expansive interpretation of how much time an officer needs to complete the mission of a traffic stop, certainly the "time needed" to handle the matter for which the traffic stop was initiated ended when Byrd handed Diaz the warning. Under Rodriguez, for any actions taken after that point, the officers needed reasonable suspicion.

2. Reasonable Suspicion

For reasonable suspicion, an officer must be able to articulate more than an "'inchoate and unparticularized suspicion or hunch' of criminal activity." Illinois v. Wardlow, 528 U.S. 119, 123-124 (2000) (quoting Terry v. Ohio, 392 U.S. 1, 27 (1968)). The reasonable suspicion test is one of totality of the circumstances. United States v. Perkins, 363 F.3d 317, 319 (4th Cir. 2004). However, the court may evaluate the "cumulative information" available to the police officer to determine if a series of innocent acts taken together can rise to the level of reasonable suspicion. United States v. McBride, 676 F.3d 385, 392 (4th Cir. 2012).

This court first surveyed how courts in this circuit had ruled in similar factual situations post-Rodriguez to set forth some guideposts by which this court could determine if Byrd impermissibly extended the duration of the traffic stop to conduct the dog sniff. Namely, the court uses two post-Rodriguez Fourth Circuit cases to draw out some factors that are important in determining whether reasonable suspicion existed, one in which the Fourth Circuit found that there was reasonable suspicion for a dog sniff and another where there was not.

In United States v. Williams, 808 F.3d 238, 244 (4th Cir. 2015), the Fourth Circuit found that there was no reasonable suspicion to deploy a drug dog, even viewing the evidence in the light most favorable to the government as required by United States v. Watson, 703 F.3d 684, 689 (4th Cir. 2013). The government's argument for a finding of reasonable suspicion rested on five factors: (1) the defendants were traveling in a rental car; (2) the defendants were traveling "on a known drug corridor at 12:37 A.M."; (3) defendants' stated travel plans were inconsistent with the due date for the return of the rental car; and (4) the defendant was unable to provide a permanent home address in New York even though he claimed he lived there part-time.¹⁰ The Williams court first dismissed the factors that defendant was using a rental car and that defendant was driving at night on an interstate highway as constituting reasonable suspicion, finding that "the overwhelming majority of rental car drivers . . . are innocent travelers with entirely legitimate purposes" and that there was "no basis on this record for assigning some nefarious significance to the 12:37 AM time of traffic stop," as "it is far from self-evident that interstate trafficking of drugs and contraband is more common at night." Williams, 808 F.3d at 249. For the third factor, that of the expiring rental agreement, the Williams court quoted the Tenth Circuit's decision in United States v. Santos, 403 F.3d 1120, 1129 (10th Cir. 2005) that "law-abiding rental car users frequently extend the rental without incurring a penalty or paying a higher rate" and that a "large number of innocent travelers [] extend their trips beyond the time originally provided for in their rental agreements" in finding that just because the defendant's travel plans exceeded the duration of the rental agreement there was no reasonable suspicion of criminality. Williams, 808 F.3d at 249. Finally, the Fourth Circuit rejected the fourth factor, that the defendant was unable to provide a home address.

10 The court notes that in Williams, the drug dog was already on the scene at the time that the vehicle was stopped initially. Id. Here, Rao arrived on the scene with Sweatman while Byrd was giving Diaz the warnings. Gov.'s Resp. 4. This means that at some point during the traffic stop, Byrd called for Sweatman and Rao to come to the scene. Calling a drug dog is not within the mission of a traffic stop. This fact bolsters the court's conclusion that the traffic stop had been impermissibly extended by the drug dog sniff and so the officers needed reasonable suspicion to conduct the free-air dog sniff. It follows that the applicable rule is that of Rodriguez as opposed to the rule in Caballes.

More recently, the Fourth Circuit in United States v. Palmer, 820 F.3d 640, 647 (4th Cir. 2016) upheld a district court's denial of a motion to suppress where the following reasonable suspicion factors existed to extend the traffic stop with a drug dog sniff: (1) defendant was in a high crime area where citizens were complaining about drug dealing; (2) the officer believed that the car's windows were illegally tinted; (3) the defendant was nervous during the traffic stop; (4) the car emitted an "overwhelming" scent of air freshener from the multiple air fresheners in the car; (5) defendant was a suspected member of a violent gang; (6) defendant's driver's license listed a P.O. box address rather than a residence; (7) defendant was driving a vehicle registered in another person's name; and (8) defendant had a criminal record that included four previous arrests for narcotics charges as well as a charge of possession of a firearm by a convicted felon. The Palmer court found it particularly "compelling" that when the officer approached the car he smelled an overwhelming odor from the air fresheners that he could see in the vehicle, suggesting that the defendant was attempting to mask the smell of drugs. See also United States v. Foreman, 369 F.3d 776, 785 (4th Cir. 2004) (concluding that air fresheners on rearview mirror supported reasonable suspicion because they are "commonly used to mask the smell of narcotics").

In considering whether the factors articulated by a police officer amount to reasonable suspicion, the court is to "separately address each of these factors before evaluating them together with the other circumstances of the traffic stop." United States v. Powell, 666 F.3d 180, 187-88 (4th Cir. 2011). In Williams, the court conducted the "totality of the circumstances" analysis by first reviewing each of the facts relied upon by the district court during an evidentiary hearing on a suppression motion first separately and then in the aggregate. This analysis means that the court should first parse out each of the factors that the officers believe constituted "reasonable suspicion" to prolong the traffic stop to conduct the drug dog sniff separately to determine if each factor rises to the level of reasonable suspicion, and eliminate those factors that are "innocent facts" when analyzed separately. For example, the fact that Diaz was driving a rental car and that he was driving on a major interstate highway after midnight are "innocent facts" as discussed by the Williams court. After eliminating those innocent facts, the court must conduct a "totality of the circumstances" analysis with the remaining facts to determine if those taken together could constitute reasonable suspicion. The ultimate inquiry is whether the factors "taken together . . . eliminate a substantial portion of innocent factors." Williams, 808 F.3d at 251.

In its briefing, the government's totality of the circumstances analysis consists of one sentence—that the officers had reasonable suspicion to prolong the traffic stop because "a driver and passenger of a vehicle rented less than a week earlier and wreaking [sic] of cologne and air fresheners cannot keep their stories straight as to whose wedding they attended." Gov.'s Resp. 6. During the hearing, the government brought out more factors, specifically that: (1) Byrd noticed a strong odor of air freshener and cologne in the car, with five air fresheners in view including in the front air vents, rearview vents, the rearview mirror, and the rear passenger area; (2) the time of the traffic stop was after midnight; (3) the "nervous demeanor" of the defendants; (4) contradictory answers from Diaz and Bozzetti about whether the wedding in Pennsylvania occurred on Saturday or Sunday and whose wedding they had been attending; (5) the rental agreement was expired, as it showed that the car was to be returned to Miami by June 3rd at 3 PM, which was approximately 36 hours prior to the traffic stop.¹¹ The court now turns to consider these factors in greater depth.

11 Neither English nor Sweatman wrote a substantive report in this case. The only report was written by Byrd. Tr. 23:21-25.

1. Air Fresheners

The first factor in the reasonable suspicion analysis is the number of air fresheners in the car. Specifically, Byrd testified that as he approached the car, he was "hit with the smell" of the air fresheners, as well as "probably like a cologne smell." Tr. 165:5-12. Byrd testified that he visually observed five air fresheners in the car. Tr. 165:22-23. He also observed newly purchased cologne, with "boxes all well" in the car, which he testified in his experience was often used as a "masking agent" to mask the odor of drugs. Tr. 166:2-6; 18-24. English testified that there was "an overwhelming smell of air freshener and cologne coming out of the car." Tr. 12:21-23. English visually observed the air fresheners, "several hanging from the rearview mirror" and one "hanging from the rear passenger door handle," which English further testified was "a little bit unusual" in a "new rental car" and in his experience was "normally" to mask the odor of narcotics "from either the officer and/or the canine." Tr. 13:7-13.

Certainly, courts have found the "overwhelming odor from the air fresheners that he could see in the vehicle" as supporting an officer's reasonable suspicion that there are drugs in the car. For example, in United States v. Foreman, 369 F.3d 776, 785 (4th Cir. 2004) the court concluded that air fresheners on rearview mirror supported reasonable suspicion because they are "commonly used to mask the smell of narcotics." In United States v. Palmer, 820 F.3d 640, 652 (4th Cir. 2016) the court found it "compelling" that an officer who approached a "darkly tinted Nissan' in a "high-crime area" where police had recently received complaints about illegal drug activity smelled the "overwhelming odor from the air fresheners." Of course, there the air fresheners were just one "concrete factor" in the reasonable suspicion analysis, which in this particular situation included the officer's knowledge that Palmer was a suspected member of a gang which was frequently involved in narcotics distribution, as well as Palmer's four earlier arrests on drug charges. Id. Here, Diaz was stopped on an interstate highway in a rented Nissan with no tinted windows, and the officers had no knowledge of any gang affiliation or of any previous convictions. Furthermore, three of the five air fresheners were clip-in air fresheners, which the Seventh Circuit recently discussed in Rodriguez-Escalera as "three or four small sticks clipped into the car's air vents" which were "not 'excessive' to the point of suggesting unlawful activity." United States v. Rodriguez-Escalera, — F.3d —, 2018 WL 1178359, at *6 (7th Cir. Mar. 7, 2018).

And finally, a review of the dashcam video shows Diaz explaining to the officers that the reason why there were so many air fresheners in the car was because all three men in the car were smoking cigarettes and the rental car agreement prohibited cigarette smoking in the car. See also Tr. 60:10-21 (English testifying that Diaz explained to Byrd that they were not allowed to smoke in the rental car). Indeed, when asked if there was any visual evidence in the car that defendants were "heavy cigarette smokers," English admitted that "[t]here was numerous packs of Marlboros . . . cigarettes throughout the vehicle." Tr. 65:15-18. The court may consider how facts later obtained "mitigate or dispel" suspicion. Rodriguez-Escalera, 2018 WL 1178359 at *6 (citing Terry, 392 U.S. at 28). The court in United States v. Wrobel, 2018 WL 297570, at *6 (D. Idaho Jan. 3, 2018) confronted a similar scenario. There too, the officer testified that he suspected the defendant was attempting to cover up the smell of narcotics his vehicle because there was a strong odor of air freshener in the car. Id. In addition to the smell of air freshener, the officer also observed an ash tray and cigarette butts in the center console of the vehicle. Id. The Wrobel court reasoned that "[r]ental companies typically prohibit smoking in rental cars, and charge substantial cleaning fees when renters smoke in their vehicles" and that therefore it was not "particularly suspicious for a driver to obtain and use air freshener when smoking cigarettes in a rental car." Id. Given that Diaz provided an explanation of the odor that the air fresheners were used to mask—cigarette smoke—and the officers testified that they visually observed packs of cigarettes, the court gives minimal weight to the "overwhelming odor" of air fresheners in the rented Nissan.

2. Source Cities

Next, Byrd and English both testified that Diaz was traveling to Miami, a "known source city." English testified that he found it suspicious that defendants were "traveling in the middle of the night" in a rental car "from one source city"—Philadelphia—"to another"—Miami. Tr. 20:14-18. He went on to explain his reasoning that "we have subjects that are traveling in the middle of the night in a rental car from one source city back to another." Tr. 20:10-18. Byrd testified that in his experience working as a trooper on I-95, "Miami is a source city for things like drugs," and "people coming out of Miami are known for carrying that thing, because it's a source city right there by the port." Tr. 144:15-16.

As the Fourth Circuit noted as early as 1991, "the vast number of persons coming from those 'source cities' relegates this factor to a relatively insignificant role." United States v. Wilson, 953 F.2d 116, 125 (4th Cir. 1991); see also Reid v. Georgia, 448 U.S. 438, 441 (1980) (holding that defendant's arrival from "source city" was an insufficient foundation for reasonable suspicion and stating the "circumstances describe a very large category of presumably innocent travelers"). The court echoes Chief Judge Gregory's concurrence in United States v. Foreman, 369 F.3d 776, 795 (4th Cir. 2004) that "given the ubiquity of 'source cities,' it seems that every traveler on an interstate road in the United States is likely headed in the direction of, or from, a 'source city' along that road, ostensibly placing them in a 'drug corridor.'" And indeed, when pressed Byrd conceded that while any large city would not be considered a source city "as much as" Miami, cities such as Atlanta and New York could all be considered source cities. Tr. 215:14-216:17. Therefore, the court gives minimal weight to the fact that the car was traveling to a source city.

3. Rental Car

Byrd stated that a number of people transporting drugs on interstate highways do so in rental cars. Certainly, "illegal transport of drugs often involves the use of rental cars traveling from source cities such as Miami." United States v. Thomas, 913 F.2d 1111, 1116 (4th Cir. 1990). However, as the Fourth Circuit explained in United States v. Williams, 808 F.3d 238, 247 (4th Cir. 2015), the use of a rental car is of "minimal value to the reasonable-suspicion evaluation" because the "overwhelming majority of rental car drivers on our nation's highways are innocent travelers with entirely legitimate purposes." Therefore, the use of a rental car is of minimal value to the reasonable suspicion analysis.

4. Interstate Highways

Byrd also explained that interstate highways such as I-95 can be drug corridors. But as the Williams court made clear, "many drug traffickers use interstate highways such as I-85, but so do many more innocent motorists." United States v. Williams, 808 F.3d 238, 248 (4th Cir. 2015). Standing alone, the observation that Diaz used an interstate highway is entitled to little weight in the reasonable suspicion analysis.

5. Nervous Demeanor

Both Byrd and English testified at length about the defendants' nervous demeanor. Byrd testified that Diaz was "extremely nervous" and his body language included "[l]ip quivering, overdramatic with the hand gestures," and that Diaz "overall looked like he was scared for some reason." Tr. 145:19-25. Before English approached the car himself, he was informed by Byrd that he thought Diaz "had more to hide than just being tired or in a hurry to get back to Florida" because Diaz "wouldn't make eye contact" and his bottom lip was "quivering." Tr. 10:16-20. English testified that when he spoke to Diaz, Diaz "wouldn't make eye contact," and when he gave responses "would sort of look out across the median" which English testified was "normally a sign of deception." Tr. 12:1-4.

Byrd and English's observations about Diaz's nervous demeanor are rebutted by the dashboard video in this case, which shows just how many times Diaz makes eye contact with Byrd. Certainly, the video of the traffic stop is "the most reliable evidence of the detainees' demeanor" as the Seventh Circuit recently recognized in United States v. Rodriguez-Escalera, — F.3d —, 2018 WL 1178359, at *5 (7th Cir. Mar. 7, 2018). The court need not credit the officers' testimony that Diaz was "extremely nervous," Tr. 10:5, where it is apparent from the dashcam video that Diaz calmly answered all of Byrd's questions while standing outside on a noisy and busy interstate highway at a late hour of the night. See Rodriguez-Escalera, 2018 WL 1178359, at *5 (Holding that the court was not required to credit an officer's testimony that a defendant "appeared nervous" where the court's own review of the video footage of the traffic stop led it to determine that the defendant was not acting "suspiciously nervous.").

Furthermore, as the Fourth Circuit cited with approval in its most recent foray into the realm of dog sniffs in United States v. Bowman, — F.3d —, 2018 WL 1093942 (4th Cir. Mar. 1, 2018), nervousness "is an unreliable indicator, especially in the context of a traffic stop." The Bowman court reiterated that the Fourth Circuit has recognized "on multiple occasions" that a driver's nervousness "is not a particularly good indicator of criminal activity, because most everyone is nervous when interacting with the police." Bowman, 2018 WL 1093942, at *8, citing United States v. Palmer, 820 F.3d 640, 651 (4th Cir. 2016). The Bowman court distinguished "nervous, evasive behavior," such as the suspect who fled upon seeing the police in Illinois v. Wardlow, 528 U.S. 119, 123 (2000). As the dashcam video reveals, Diaz was calm and cooperative during the traffic stop. At no point did he attempt any evasive behavior. As is present here, signs of nervousness, standing alone, are to be discounted when determining the officer's basis for reasonable suspicion. The Bowman court reasoned that "[t]here is nothing intrinsically suspicious or nefarious about the occupant of a vehicle not making eye contact with an officer during a traffic stop." Id. at *9. It went on to state that "[g]iven the complex reality of citizen-police relationships . . . a young man's keeping his eyes down during a police encounter seems just as likely to be a show of respect and an attempt to avoid confrontation." Id. (citing United States v. Massenburg, 654 F.3d 480, 488 (4th Cir. 2011)). The court reiterates this point—an officer's recitation of "furtive movements" is not talismanic in the reasonable suspicion analysis. Therefore, the court does not give this factor much weight in its reasonable suspicion analysis.

6. Expiring Rental Agreement

Byrd said that an additional factor to make him develop reasonable suspicion was that Diaz was driving a rental car, where a review of the rental car agreement demonstrated that it had expired hours before the traffic stop. Tr. 151:15-18 (Byrd testifying that "it seemed strange" when he was "looking at a rental agreement that's expired" because "that costs a lot of money," and that the expired rental agreement was "another little bullet point that [he] noticed that was strange.") Byrd observed that the rental agreement was expired at the beginning of the stop, when Diaz handed it over, but did not ask Diaz about it "at that time." Tr. 148:7-9. "[I]nnocent travelers frequently extend rental agreements." United States v. Williams, 808 F.3d 238, 250 (4th Cir. 2015). And indeed, when Byrd questioned Diaz about his expired rental agreement Diaz answered that he had called the rental car company to extend the rental agreement. Tr. 163:10-14. This factor is of limited value in the court's reasonable suspicion analysis.

7. Conflicting Stories of Wedding

Finally, Diaz and Bozzetti gave conflicting stories about whose wedding they were going to attend and the date of the wedding. Diaz informed Byrd that they were traveling from Pennsylvania, and they had traveled to Pennsylvania for Diaz's "brother's wedding that occurred on Sunday." Tr. 143:17-20. After speaking with Diaz, Byrd then went to speak with Bozzetti. Tr. 149:1-25. Bozzetti also told Byrd that they had been travelling from Pennsylvania, but said that it was to attend Diaz's sister's wedding, and that the wedding occurred on Saturday. Tr. 149:8-10; Tr. 150:22-23. English testified that Byrd told him about the discrepancies in the stories of the passengers about the date of the wedding they were attending and whose wedding it was. Tr. 10:3-11. English then went to speak with Diaz and Bozzetti, who repeated their stories to him. Tr. 164:1-7. This was, according to English, "just a lot of conflicting stories for three men that were traveling together." Tr. 10:11-13. English reasoned that "if they had actually done these things [they] shouldn't have any problem recalling their story." Tr. 10:13-15.

The question of whether inconsistent statements constitute reasonable suspicion to extend a traffic stop is not one that has been squarely confronted by the Fourth Circuit, although the cases largely support the government's argument that inconsistencies can, in conjunction with other factors, constitute reasonable suspicion. In United States v. Sang, 89 F.3d 831 (4th Cir. 1996) (per curiam) (unpublished table decision), the Fourth Circuit found reasonable suspicion where a driver's stated travel plan was inconsistent with the travel plan provided by the passenger.

But the factor of inconsistent stories in formulating reasonable suspicion has been discussed at some length by other courts. A line of cases from the Fifth Circuit supports defendants' argument that inconsistent stories are insufficient to create reasonable suspicion. In United States v. Estrada, 459 F.3d 627, 631 (5th Cir. 2006), the court held that reasonable suspicion "does not arise solely from the inconsistent answers regarding the length of . . . ownership of the truck," although it ultimately held that there was "significant physical evidence" that there was a hidden compartment in the truck where illegal narcotics were ultimately found that supported reasonable suspicion. In United States v. Santiago, 310 F.3d 336, 338-39 (5th Cir. 2002), the court recognized that inconsistent stories between a driver and a passenger about the length of their planned trip and "uneasy feelings" about the passengers' demeanor did not support reasonable suspicion of drug trafficking. In the Sixth Circuit, both United States v. Orsolini, 300 F.3d 724, 726 (6th Cir. 2002) and United States v. Hill, 195 F.3d 258 (6th Cir. 1999) are instructive in guiding the court about when inconsistent stories in conjunction with other factors are enough for reasonable suspicion. In Orsolini, the driver and passenger gave inconsistent stories about whom they were going to visit. Orsolini, 300 F.3d at 728. The Sixth Circuit found that the inconsistent stories, in conjunction with other factors such as the recent purchase of the vehicle with cash in a source city for drugs and that the driver's only proof of identification was a photocopy of an interim license, were sufficient to support reasonable suspicion. Id. at 729. In Hill, the driver and passenger gave inconsistant stories regarding their itinerary. Hill, 195 F.3d at 271. The Sixth Circuit held that these inconsistent stories, in conjunction with the defendants' implausible explanation for the trip and the nervous behavior exhibited throughout the stop was sufficient for reasonable suspicion. Id. at 272. But in United States v. Mesa, 62 F.3d 159, 162-63 (6th Cir. 1995), the Sixth Circuit rejected an officer's argument that he had reasonable suspicion where the passengers told inconsistent stories regarding whether the relative they had visited had a heart attack or a stroke.

Here, the inconsistencies in the stories of the passengers of whose wedding the defendants had attended in Pennsylvania and on what day the wedding took place certainly supported the conclusion that one or more of the passengers were lying about the reason for their trip. Furthermore, the inconsistent stories between Diaz and Bozzetti on the day that the wedding in Pennsylvania occurred and whose wedding had taken place was not the only circumstance that created reasonable suspicion in Byrd's mind. But Byrd presented no testimony about why the alleged inconsistencies led him to believe that the defendants were involved in criminal activity, just that they gave him inconsistent stories about their trip.

Weighing the testimony of the officers and the video and audio recording of the traffic stop, the court concludes that there was no reasonable suspicion to allow Byrd to extend the stop. Contrary to the officer's description of Diaz's nervous demeanor, a review of the dashcam video reveals that Diaz was relatively calm and collected throughout the stop. The court also concludes that the air fresheners that the officer observed in the Nissan were not so excessive as to raise reasonable suspicion, especially where Diaz informed the officers that the air fresheners were to mask the odor of cigarette smoking which was prohibited by the rental car agreement and officers observed empty cigarette packs on the floor of the vehicle. No criminal history, tips, or surveillance supported Byrd's suspicions about Diaz, who testified that he had no knowledge of the defendants before initiating the traffic stop. Tr. 162:7-9. Cf. United States v. Finke, 85 F.3d 1275, 1282 (7th Cir. 1996) (criminal history check revealed that defendant had two prior drug convictions, which "strongly confirmed [the police officer's] initial suspicions" about narcotics activity). Certainly, the conflicting stories of whose wedding was in Pennsylvania and the date of the wedding could support reasonable suspicion, but the entire set of circumstances taken together do not rise to the level of reasonable suspicion.

This court recognizes and gives appropriate weight to the officers' assessment of certain circumstances based on their training and experience.¹² However, it also reiterates the warning of courts before it, that the government cannot "simply proffer 'whatever facts are present, no matter how innocent, as indicia of suspicious activity.'" United States v. Foster, 634 F.3d 243, 248 (4th Cir. 2011). And like courts before it, this court has concerns about "the way in which the Government attempts to spin . . . mundane acts into a web of deception." Id. The court therefore holds that under Rodriguez, there was no reasonable suspicion to extend an otherwise-completed traffic stop such that the dog sniff could be executed. Defendants' motion to suppress is granted on this ground alone. And as explained below, even if the officers had reasonable suspicion to prolong the traffic stop, Rao was not reliable such that its positive "alert" constituted probable cause to conduct the search of the car.

12 Defendants brought up the "serious question" of "whether or not [the officers] were delaying this traffic stop in order to develop probable cause or to develop reasonable suspicion." Tr. 243:20-22. The court shares the defendants' concern with how the officers' involvement in the ACE team influenced their reasonable suspicion calculus. The ACE team is trained in "proactive traffic enforcement methods," and are instructed to "investigate beyond the traffic summons or warning" when "reasonable suspicion is developed." Def.'s Ex. 3, South Carolina Highway Patrol Aggressive Criminal Enforcement Interdiction Team Principles. This ACE team was tasked with, according to English, "looking for felons, drugs, things beyond the normal traffic stop." Tr. 7:23-25. English elaborated that as part of the ACE team, he was trained to "[g]o beyond the license plate," and that he in turn taught classes on "things that people may say or do or things that you may see in a car that makes you want to ask a few more question[s] and develop some reasonable suspicion or probable cause." Tr. 8:1-13. English conceded in his testimony that as a part of the ACE team, he was "trained to develop reasonable suspicion." Tr. 24:21-25. This is "developed," according to English, through interviews with passengers during traffic stops, because "[i]f you don't have any conversation with them whatsoever, then there's no way to develop that probable cause." Tr. 26:1-4. Sweatman concurred that as a part of the "ACE team," he was trained "to develop probable cause or reasonable suspicion." Tr. 101:12-14. English further explained that when Byrd called for backup during the traffic stop, he called for "an additional R unit, which is specific to the ACE team, as "typical troopers" are referred to as "F units." Tr. 8:16-20. The composition of the ACE team unit that patrolled I-95 was Sweatman, English, and Byrd, all of whom had drug dogs in their respective cars. Tr. 9:10-14.
While the court makes no finding, it does note that the structure of the ACE team seems to condition officers to create reasonable suspicion where there may be none. This demonstrates the wisdom of the observation that "a man hears what he wants to hear, and disregards the rest." N. Indiana Pub. Serv. Co. v. Colorado Westmoreland, Inc., 667 F. Supp. 613, 631 (N.D. Ind. 1987), aff'd, 845 F.2d 1024 (7th Cir. 1988) (citing Paul Simon, "The Boxer" (1968)).

D. Dog and Alert Reliability

After Diaz refused consent to search the car, Sweatman arrived and ran Rao around the car. Tr. 14:1-5. Rao then "alerted" to the "driver's door area" of the car, according to English. Tr. 14:21-22. Defendants also move to suppress the evidence on the ground that Rao and the alert itself were not reliable enough to establish probable cause to search the car. The court agrees on both grounds.

1. Dog Alert

Defendants first challenge the determination that the drug dog alerted. Whether a dog alerts is a question of fact for this court to resolve. United States v. Mason, 628 F.3d 123, 130 (4th Cir. 2010). It is undisputed that it was based solely on Rao's "alert" that the officers believed they had probable cause to search the car. Tr. 15:11-13. Sweatman conceded that there were no drugs in the car, and neither was there any paraphernalia from which to deduce that drugs were or had been in the car such as a pipe or sandwich baggies. Tr. 100:15-101:3.

What constitutes an "alert" is not clear from the caselaw. Some courts have adopted the view that an alert is trained behavior that can be either aggressive, such as biting or scratching, or passive, such as sitting or looking. See United States v. Johnson, 323 F.3d 566, 567 (7th Cir. 2003). By contrast, others have determined that a change in behavior is insufficient to constitute an alert and instead that the drug dog must perform the full indication that it is trained to do before it constitutes an alert such that the officers have probable cause to search the vehicle. See United States v. Rivas, 157 F.3d 364 (5th Cir. 1998). The Fourth Circuit has not yet determined what the proper "alert" standard is for a drug detection dog.

The difference between an "alert" and an "indication" is important to the court's analysis on whether Rao alerted. Sweatman testified that only he, as Rao's handler, would be able to determine if Rao "alerted," as "[e]ach situation would vary" in determining whether Rao alerted. Tr. 111:1-12. Sweatman extrapolated on this difference between an alert and an indication, explaining that an alert "is a behavior exhibited . . . interpreted by the handler of the canine . . . typically a change in body posture, a change in rapid respiration." Tr. 108:20-25. In contrast, Sweatman testified that an indication "is a trained behavior for the canine to exhibit when he believes he has reached as close as he possible can to what he believes the source of the trained odor is." Tr. 109:3-5. An indication is distinct from an alert, according to Sweatman, as it is actually trained by the trainer of the dog and simply "maintained" by the handler. Tr. 111:20-22.

As best the court can discern, the government is arguing that Rao showed alert behavior by a change in rapid respiration and change in body posture but did not indicate—but that this "alert behavior," which only Sweatman as Rao's handler could discern constitutes a dog alert. A review of the dashcam video reveals only that Rao jumped up and down while Sweatman tries to restrain it. The court has no information by which it can determine that this is how Rao was trained to "alert"—indeed, the court understands that Rao was trained only to "indicate," and finds that he did not show this trained "indication" behavior in the dashcam video.

The parties did not call any expert witnesses to testify on what constitutes a dog alert, and the government now urges the court to accept Sweatman's testimony as conclusive that Rao alerted. Certainly, just because the government did not offer an expert witness does not mean that Rao is not reliable. The Fourth Circuit has made clear that there is no requirement that dog alert testimony needs to satisfy the requirements for expert scientific testimony, as the dog alert is not actual evidence of drugs but "establish[es] probable cause to obtain a warrant to search for such substantive evidence." United States v. Allen, 159 F.3d 832, 839-40 (4th Cir. 1998). But what concerns the court is that it had only the law enforcement officers' testimony to consider in evaluating whether Rao did in fact "alert." And of the three officers involved in the stop, Byrd testified that he did not personally observe Rao "indicate" or "alert," as he was "basically watching the occupants of the car" and testified only that Sweatman advised him that Rao "did indicate [as opposed to alert] on the driver's side of the vehicle." Tr. 169:18-21.

During the suppression hearing, English, who had a drug dog himself, testified that he observed Rao "alert," which he defined as "a change in behavior" such as "[f]aster respirations" or a "change" in the dog's posture. Tr. 14:21-15:9. English further testified that the handler "needs to know" how the drug dog reacts, but that the drug dog is trained to react in a certain manner. While this training is done before the drug dogs arrive at the SCHP—ostensibly at the Alabama Canine Law Enforcement Training Center ("Alabama Canine Center") where the drug dogs are trained—the court was presented with no records of how Rao was trained to indicate. Sweatman testified that an "alert" "is simply a behavior as perceived by the handler of that particular canine. An alert behavior could be different for each individual dog, but an alert for a dog is perceived by that handler." Tr. 86:21-24. Specifically, Sweatman testified that Rao's "alert" was when he would "drop posture," and that "[t]ypically his ears may move backwards," and he would "typically pull me toward where he's picking up trained odor." Tr. 87:3-8. Sweatman went on to say that Rao could alert by "you know, leap[ing] up on the side of something." Tr. 87:11-14. In this particular search, Sweatman testified that he knew Rao alerted because it "changed to rapid respiration, change in body posture." Tr. 92:14-25. Sweatman testified that Rao's "trained behavior" is that it is "trained passively, to indicate passively where he would either sit, lay, or stare." Tr. 94:6-12. But when asked to pinpoint when in the dashcam video Rao "sat and indicated," Sweatman admits that "you can't get a clear—a clear visual of that on this footage." Tr. 96:1-25.

The court in United States v. Wilson, 995 F. Supp. 2d 455 (W.D.N.C. 2014) confronted a similar set of facts. In Wilson, the drug dog's trained indication was to "sit and stare" when it detected narcotics. Id. at 474. The court in Wilson reviewed the dashcam video and found that at no time did the dog "sit and stare." Id. The government then argued that the court should accept the handler's testimony that although the dashcam video did not show that the drug dog "sat and stare" it had demonstrated other behavior that the handler interpreted to mean that the drug dog had "casted" or "alerted" to the oxycodone—which the dog's handler admitted that it was not trained to detect. Id. at 475. The Wilson court refused to accept such testimony from a drug dog handler, reasoning that "with law enforcement's use of detection dogs, the dog becomes the functional equivalent of the magistrate," and so it was reasonable to expect "the objective basis informing the probable cause determination" in such a case. Id. The Wilson court reasoned that to accept the handler's "subjective determination that a dog has made some otherwise undetectable alert" was untenable, as such a conclusion would be "for all practical purposes, immune from review" and ultimately found that the drug dog's "'change in behavior' cannot be equated to an 'alert.'" Id. The Wilson court ended with the following caution:

To allow a search predicated upon an officer's interpretation of the utterly minimalist lesser showing exhibited by the dog in this case would be tantamount to permitting law enforcement officers to issue

their own search warrants based upon their own subjective analysis, something the Framers explicitly prohibited.

Id. This court is guided by the Wilson court's well-reasoned opinion, and refuses to accept Sweatman's testimony that the dog alerted for those reasons.

Based on the evidence submitted to the court during the evidentiary hearing and after considering the relevant case law the court finds that Rao's behavior change did not constitute an alert.

2. Drug dog reliability

Although the court has already found that Rao did not "alert," the court now turns to analyze the reliability of Rao itself as an alternate ground for granting the motion to suppress. It is well-established that a "reliable" dog alert establishes probable cause that illegal drugs are present. Illinois v. Caballes, 543 U.S. 405, 409-10 (2005). In Harris, 568 U.S. at 244, the Court found that to assess the reliability of a drug dog the government need not create a strict evidentiary checklist such as introducing "comprehensive documentation of the dog's prior 'hits' and 'misses' in the field," but that a finding of "reliability" for a drug dog should be a balancing test.¹³ The Harris court found that "evidence of a dog's satisfactory performance in a certification or training program can itself provide sufficient reason to trust his alert." If the dog does not have a formal certification it has undergone to test its reliability in a controlled setting, the dog could still be reliable if "the dog has recently and successfully completed a training program that evaluated his proficiency in locating drugs."

13 The government discussed Harris briefly at the conclusion of the second suppression hearing, stating "I was going to address Harris. . . but my understanding is that case has largely been treated as an outlier in this particular type of situation." Tr. 249:10-14. The court questions whether any decision of the Supreme Court could be described as an "outlier" and the government offered no evidence that Harris has been considered an "outlier" where the reliability of a drug detection dog is challenged. This court follows Harris as the seminal case on the standard of probable cause in determining reliability of a drug dog.

The Harris court points out that during a probable cause hearing focusing on the dog alert, the court should evaluate the proffered evidence to determine what all of the circumstances demonstrate. Harris, 568 U.S. at 248. If defendants have challenged the government's case by "disputing the reliability of the dog overall or of a particular alert," then the court should weigh the competing evidence. The Harris court discusses that even a drug dog that is "generally reliable"—for example, through evidence of the dog's certification by a "bona fide organization"—the circumstances surrounding the particular, challenged alert could undermine the case for probable cause. This could include if the officer "consciously or not" cued the dog, or if the team was "working under unfamiliar conditions." Certainly, the factors offered by the Harris court are not exhaustive, but they offer a roadmap of how the court should conduct its probable cause analysis on the reliability of the drug dog both generally as a drug dog and under the specific circumstances of the search of Diaz's car.

a. False Positives

The defendants point out that the dog sniff was a false positive,¹⁴ as no drugs were found in the car.¹⁵ This is certainly true—Rao was not trained to detect credit cards. And indeed, Sweatman admitted that it was "not uncommon" for Rao to alert or indicate but drugs were not found in the vehicle. Tr. 112:6-14.

14 Certainly, the rate at which the drug dog detected false positives is just another factor in the court's determination of whether there was probable cause to search the car. Here, the dog alert was not the only evidence that the government could have used to argue that the officers had probable cause to search the car. In addition to the dog's positive alert, there was the conflicting information given by the car's occupants on when the sister's wedding was, the furtive movements of the passenger in the passenger side of the car, the passenger's response to questioning regarding contraband, and the air fresheners in the car. Byrd testified that based on his training and experience, circumstances such as the air fresheners in the car and Diaz's nervous demeanor are indicators of criminal behavior. But as discussed at length in Section III.2, the court does not find that these circumstances rose to reasonable suspicion such that Byrd could have prolonged the traffic stop to conduct the dog sniff, let alone to the level of probable cause to allow the search of the car.

15 The court notes that Rao was not trained to detect drugs but to detect the odor of drugs. As the Harris court explains:

A detection dog recognizes an odor, not a drug, and should alert whenever the scent is present, even if the substance is gone (just as a police officer's much inferior nose detects the odor of marijuana for some time after a joint has been smoked). In the usual case, the mere chance that the substance might no longer be at the location does not matter; a well-trained dog's alert establishes a fair probability—all that is required for probable cause—that either drugs or evidence of a drug crime . . . will be found.

However, the Fourth Circuit has adopted the reasoning of the Eighth Circuit in finding that "a low percentage of false positives is not fatal to the finding that a drug detection dog is properly trained and certified." United States v. Age, 447 F. App'x 470, 471 (4th Cir. 2011) (citing United States v. Scott, 610 F.3d 1009, 1014 (8th Cir. 2010), cert. denied, 562 U.S. 1160, (2011)). In Age, the Fourth Circuit found that there was sufficient evidence of a dog's reliability where the government provided evidence regarding the dog's detailed training and continuing certification, the officer testified that he had worked with the dog for years, and that the dog's alerts were 80-90% accurate. Exactly what percentage of false positives constitutes a "low percentage" is not immediately discernible from the caselaw. In United States v. Donnelly, 475 F.3d 946, 955 (8th Cir. 2007), which the Age court cited, the Eighth Circuit held that even a 54% accuracy rating for a drug dog was sufficient to find probable cause "taking into account the totality of the circumstances present at the scene . . . , [the defendant's] behavior and condition, [the dog's] history and pedigree, and [the dog's] positive indication of drugs within the vehicle." In United States v. Koon Chung Wu, 217 F. App'x 240 (4th Cir. Feb. 2, 2007), the Fourth Circuit discussed the accuracy rate of Cody, a drug dog:

The district court found that Cody had demonstrated an impressive degree of accuracy in training exercises, and based on its review of Cody's past field sniffs, the court found that Cody was accurate approximately 67% of the time in the field when sniffing for narcotics. While we believe that this factual finding was unnecessary for the district court to have concluded that probable cause existed for the searches given the evidence of Cody's training and certification, it serves to bolster the court's determination that Cody was sufficiently reliable for his positive alerts to establish probable cause for the May 9 and June 27 searches. See United States v. Robinson, 390 F.3d 853, 874 (6th Cir. 2004) ("[A]fter it is shown that the dog is certified, all other evidence relating to his accuracy goes only to the credibility of the testimony, not to the dog's qualifications." (internal quotation marks omitted)).

Here, the government submitted Rao's field records span some hundreds of pages in the record and cover multiple years. But while the court can consider field records, these records are prone to containing false positives and false negatives. For example, during the hearing there was some consternation about how to verify Rao's detection of an odor in the field. Specifically, Sweatman was asked about cases where Rao alerts but no drugs were found. Tr. 117:8-24. Sweatman testified that in the event that no drugs were found, he would "ask the occupants" of the vehicle "[d]epending on the situation." Tr. 117:25. Where no drugs were found in the vehicle and no indication of illegal drug usage such as drug paraphernalia was found, Sweatman testified that he would "verify" if drugs had previously been in the vehicle by asking the occupants of the vehicle if they had recently been transporting narcotics. Tr. 118:8-22. It is precisely this type of situation that makes discerning an error rate in the field difficult, and why the court finds records from a controlled environment more dependable.

What is especially concerning is not necessarily Rao's rate of false positives in the field but the lack of Rao's error rate in a controlled environment. There was no testimony offered regarding the phases of the certification test that Sweatman and the drug dog had to complete to receive certification. The government did discuss "the extent of the training" received by Rao, specifically that Rao first completed the Alabama Canine Law Enforcement Officer's Training Center 160 hour "Narcotic Certification" test in 2011 with an overall score of "Above Average." Court's Ex. 2. This test consisted of ten different searches with a total of fourteen finds in a variety of environments using varying amounts of narcotics substances such as marijuana, heroin, crack cocaine, and methamphetamine. Id. Rao then participated in a 4-hour Narcotics Re-Certification Test with the same training center in 2012, 2013, 2014, 2015, and 2016, receiving an "Above Average" score at each of these annual recertification. Id. Sweatman explained that at the Alabama Canine Center, the drug dog "would need to find all the drugs in order to pass the certification." Tr. 124:3-25. Then, during the "maintenance training" that the drug dogs underwent, Sweatman explained that if Rao "miss[ed] something" it was allowed to "try again" and the handler would "try to reintroduce the dog to that area." Tr. 127:1-19. During the recertification process done at Alabama Canine Center, Sweatman explained that when a drug dog missed something on the test, the test proctor would inform the handler "there's a problem with certification" and the handler would then "have to correct that problem," after which the drug could test again at the test proctor's discretion. Tr. 128:10-23. English, who has also had drug dogs during his time at the SCHP and indeed had a drug dog with him on the night of the search, described the training that the drug dogs receive as follows:

We're sent to a four-week handler course in Alabama with our current dogs where we receive training with the dogs to handle them, and then we're certified. Each year we go back to Alabama to get recertified again, and then every month we have monthly maintenance training we do to keep the dogs' records up for their work throughout the year.

Tr. 7:14-19. English testified that the training consists of different "hides" in a controlled environment with the drug odors that the drug dog is trained to detect—specifically, marijuana, cocaine, heroin, and methamphetamine. Tr. 84:24-85:5.

But troublingly, while the government presented reports of how Rao fared in the field, it did not disclose certification records or training standards and manuals. Indeed, the government admitted that in its review of Rao's training records it was "unable to locate any records indicating training scenarios lending themselves to determining an error rate." Id. (emphases added). The government argued that Sweatman and Rao had been "above average." Tr. 237:16-20. The court then asked "[w]hat does above average mean" and presented the hypothetical of whether recertification would occur even if "he only makes mistakes 49.9 percent of the time." Tr. 21:21-22. Counsel for the government answered that he was "not sure, based off the records [he has] seen or [his] conversations," and that he ultimately "[didn't] know the answer to that." Tr. 237:23-25.

In United States v. Cedano-Arellano, 332 F.3d 568 (9th Cir. 2003), the Ninth Circuit held that Federal Rule of Criminal Procedure 16 compels the government to disclose the "handler's log," as well as "training records and score sheets, certification records, and training standards and manuals" pertaining to the dog because these materials were "crucial to [the defendant's] ability to assess the dog's reliability, a very important issue in his defense, and to conduct an effective cross-examination of the dog's handler" at the suppression hearing. Granted, Cedano-Arellano was before the court on a discovery violation and the court ultimately found only that the defendant was "hamstrung" by his inability to properly cross-examine the adequacy of the certification program. But just as a defendant is hamstrung if the government does not provide such materials where the government seeks to rely on the drug dog alert as the evidentiary basis for its search and the defendant has contested the adequacy of the certification program, this court is similarly hamstrung.

The government contends that the defendants "offered nothing to contradict or suggest that the dog was anything but reliable." Tr. 237:14-15. But defendants did dispute Rao's reliability, arguing "[t]here is no presentation . . . of an error rate through any of the training records about what exactly a passing grade is, what is above average, what is average. . . So it's our position that [the government] [has] not shown that this is a reliable dog, based upon the training records." Tr. 245:4-8. Courts rely on the dog's certification as proof of its reliability. United States v. Kennedy, 131 F.3d 1371, 1378 (10th Cir. 1997) ("[W]ith a canine, the reliability should come from the fact that the dog is trained and annually certified to perform a physical skill.") (quotation omitted). Given that the government did not present any records of Rao's performance in a controlled environment, it follows that the court has no way of determining Rao's error rate in a controlled environment.

b. Training Facility as "Bona Fide Organization"

The court now turns to the training facility at which Rao was certified to be a K9 unit, and whether it qualifies as a "bona fide" organization such that the court can presume, subject to any conflicting evidence offered, that Rao's alert provided probable cause to search. Harris, 568 U.S. at 246. Defendants challenge the Alabama Canine Center as not a "bona fide" organization, arguing that "[t]here's been no testimony about this particular training center, that it is, you know, bona fide. They have certificates, but we don't know the procedures that they use to test the dogs. And we don't know what the passing rate even is." Tr. 245:9-13.

The primary basis upon which the Supreme Court allows a dog sniff to create probable cause is the training and certification of the canine which gives confidence in the reliability of the dog's signal that contraband is present. See Harris, 133 S. Ct. at 1057 (2013) ("[E]vidence of a dog's satisfactory performance in a certification or training program can itself provide sufficient reason to trust [the dog's] alert."). Specifically, the Harris court held that there is a presumption of probable cause where there is training or certification conducted by a "bona fide" organization. Harris, 568 U.S. at 246. It did not, unfortunately, provide the contours of just what a "bona fide" organization is.

Rao was certified through the Alabama Canine Center. Tr. 84:21-25. When pressed, Sweatman admitted that to his knowledge the Alabama Canine Center is not certified by the National Association of Public Working Dogs Association ("NAPWDA"), an organization that establishes national standards for the training of both handlers and canines. Tr. 113:4-23; 114:6-11. Certainly, the court does not view NAPWPA certification as determinative of whether Alabama Canine Center is a "bona fide" organization. And indeed, it may be that there is no requirement that the highway patrol drug dogs need to be certified by NAPWPA. Tr. 122:22-123:4. But the court was provided no evidence by which to determine the adequacy of the certification program at Alabama Canine Center—its standards could be "too lax" and its methods "faulty," as prescribed by the Harris court. Harris, 568 U.S. at 247 ("A defendant, however, must have an opportunity to challenge such evidence of a dog's reliability . . . [t]he defendant, for example, may contest the adequacy of a certification or training program, perhaps asserting that its standards are too lax or its methods faulty."). The court simply has no way to know. When pressed about the training and certification of Rao, the government argued that there is "nothing suggesting" that the training center is "anything but a reliable training center." Tr. 248: 13-15. Of course, to accept this argument would mean that a court must accept the government's word that a training center is reliable even when no evidence about the center's certification is provided to the court. The court cannot countenance such a rule. To do so would be to abdicate its responsibility to Harris to evaluate a drug dog's reliability.

The court makes no finding on whether the Alabama Canine Center is a "bona fide" organization, as it does need to—even assuming that it is, as explained above, what concerns the court most is not whether the Alabama Canine Center is certified but if the Alabama Canine Center's certification of Rao is valid such that the court can consider Rao a reliable drug dog. To answer this, the court needed records of the results of Rao's certification and recertification tests as well as his error rate in a controlled environment. The government did not present any of this evidence.

Ultimately, this court must be able to evaluate Rao's performance during controlled evaluations. The court remains unaware what the passing rate is for a certification test at the Alabama Canine Center, as well as what it means for a drug dog to achieve a score of "above average" as opposed to "satisfactory" or "excellent." The court also has no failure rates for the certification test at the Alabama Canine Center. The government has failed to provide any evidence on Rao's error rate in a controlled environment or any information about the adequacy of the certification program of the Alabama Canine Center. Therefore, even assuming that Rao did "alert"—which the court finds that he did not—the court finds that Rao was not sufficiently reliable such that his positive "alert" would constitute probable cause to search defendants' vehicle as set forth in Harris. Ultimately—and unfortunately—Rao's sniff is simply not "up to snuff." Harris, 568 U.S. at 248.

III. CONCLUSION

For the reasons set forth above, the court GRANTS the motion to suppress.

AND IT IS SO ORDERED.

/s/ _________

DAVID C. NORTON

UNITED STATES DISTRICT JUDGE April 6, 2018
Charleston, South Carolina

Harris, 568 U.S. at 246.