870 So.2d 8 (2003)
Gary Alan MATHESON, Appellant,
v.
STATE of Florida, Appellee.
No. 2D00-1611.
District Court of Appeal of Florida, Second District.
August 1, 2003.
Rehearing Denied March 5, 2004.
James Marion Moorman, Public Defender, and Celene Humphries, Special Assistant Public Defender, Bartow, for Appellant.
Charles J. Crist, Jr., Attorney General, Tallahassee, and Susan M. Shanahan, Assistant Attorney General, Tampa, for Appellee.
NORTHCUTT, Judge.
Gary Alan Matheson maintains the State failed to prove that an alert by Razor, a narcotics detection dog in service to the Hillsborough County Sheriff's Office, furnished probable cause to search his vehicle. He advanced that position during his prosecution for drug offenses in a motion to suppress the contraband discovered and seized during the search. When the circuit court denied the motion, Matheson pleaded no contest to three counts of possessing a controlled substance and one count of possessing drug paraphernalia. He reserved his right to appeal the denial of his dispositive motion to suppress. We reverse.
PROCEEDINGS BELOW
At the suppression hearing it was related that in May 1999 Razor was called upon to sniff Matheson's car after deputies stopped Matheson for a traffic infraction. During the traffic stop the deputies had made what they described as a routine request for permission to search Matheson's car. Matheson had declined; hence, the deputy called for Razor's assistance.
The State offered the testimony of Razor's handler, Deputy Greco, who recounted that he arrived at the scene as another deputy was writing Matheson a traffic citation. Deputy Greco testified that he followed his normal routine by taking Razor to the driver's side door of Matheson's car and quickly walking the dog around the car in a clockwise direction. Razor did not alert on this first pass. Deputy Greco then walked Razor slowly around the car, allowing him to linger at the "seams." This time, Razor scratched and bit at the edge of the car's hatchback, which Deputy Greco recognized as Razor's alert behavior.
Deputy Greco advised his colleagues that Razor had alerted. They then entered Matheson's car and searched it. In the rear of the car they discovered a bag containing drug paraphernalia, including syringes and spoons. In the glove compartment the deputies found hydrocodone tablets, morphine tablets, and methamphetamine.
On the evening that Deputy Greco walked Razor around Matheson's car, he had been a canine handler for about twenty-one months. He testified that he had taken training both in canine patrol handling and in narcotics detection. He and Razor had been assigned to each other since both began their services in canine patrol in August 1997. Prior to Razor's sniff of Matheson's car, he had been certified to detect marijuana, cocaine and heroin. He subsequently was certified to detect methamphetamine.
On cross-examination, Deputy Greco acknowledged that he had not maintained a record of Razor's false alert rate. In fact, he often left the scene of a sniff after advising deputies that Razor had alerted, and thus never learned whether the alert had led to the discovery of contraband.
At the conclusion of Deputy Greco's testimony, the State rested. The circuit court agreed that the State had made a prima facie showing that the search of Matheson's car was supported by probable cause.
The defense then presented the testimony of Razor's trainer, Sergeant Olive. He testified that Razor completed a thirty-day course of training by the Hillsborough County Sheriff's Office in October 1997 and a one-week program under the auspices of the United States Police Canine Association in June 1998.
During questioning about the specifics of the HCSO and USPCA training regimens, Sergeant Olive testified that Razor had received no training to discourage him from alerting to "dead scents," those being residual odors of drugs that are no longer present. Sergeant Olive also confirmed that the Sheriff's Office did not maintain records of Razor's success rate. When explaining this, he maintained that it would be impossible to assess a dog's reliability "in the street" because the dog might alert on dead scents. Sergeant Olive asserted that he would not consider an alert on a dead scent to be a false alert because the dog had done what he was conditioned and certified to do, i.e., alert to the odor of contraband.
The defense submitted the expert testimony of Dr. Dan Craig, a veterinarian and animal behavior specialist whose background included extensive consultation with the United States military and other agencies regarding their detection dog programs. Dr. Craig testified that the HCSO training procedures used with Razor were too simplistic to make him reliable at detecting narcotics for six reasons. First, Razor received inadequate training for searching vehicles. Second, Razor was not trained with small quantities of drugs. Third, training officers failed to plant novel odors during Razor's training searches. Fourth, Razor was not subject to controlled negative testing, in which all objects or locations have no drugs present. Dr. Craig said that this type of testing indicates a false response rate and reveals whether the handler or the dog is guessing. He added that preventing the handler from knowing whether drugs will be present during a training exercise reveals whether the handler is consciously or unconsciously prompting the dog to alert. Dr. Craig asserted that this type of testing is essential and should be performed periodically on a random basis. Fifth, Razor was not given extinction training, which would have discouraged him from alerting to common items that are sometimes associated with drugs, such as plastic bags used for packaging. Sixth, there was no evidence that Razor's training included "stimulus generalization," which conditions a dog trained on one class of drugs to detect all drugs in that class.
Addressing Razor's USPCA certification, Dr. Craig testified that there were a number of flaws in the USPCA certification procedures that rendered this certification insufficient evidence of Razor's reliability. First, the USPCA did not perform controlled negative testing. Second, the USPCA limited the dog's search time to ten minutes, which is shorter than "real world" searches. Third, the USPCA required only a seventy percent proficiency, which Dr. Craig considered insufficient. Fourth, the USPCA failed to focus on the dog's ability to detect narcotics, but analyzed the ability of the dog and handler as a team. Therefore, according to Dr. Craig, the USPCA could not truly certify the dog's individual ability to detect narcotics. Fifth, Razor was not certified to detect methamphetamine, and his training did not prepare him to reliably detect this substance.
DISCUSSION
Under the Fourth Amendment, a law enforcement officer may not search a place within the ambit of a person's legitimate expectation of privacy unless the officer has probable cause to believe that a search of that place at that time will uncover evidence of a crime. See Pagan v. State, 830 So.2d 792, 806 (Fla.2002), cert. denied, 539 U.S. 919, 123 S.Ct. 2278, 156 L.Ed.2d 137 (2003). Whether applying for a search warrant beforehand or justifying a warrantless search after the fact, it is the State's burden to show that the search will be or was justified by probable cause. See Doorbal v. State, 837 So.2d 940, 952 (Fla.), cert. denied, ___ U.S. ___, 123 S.Ct. 2647, 156 L.Ed.2d 663 (2003); Doctor v. State, 596 So.2d 442, 445 (Fla.1992).
In this case the State contends that it met its burden based on the testimony of Deputy Greco. It maintains that by proving Razor was trained and certified, it established prima facie that Razor's alert gave the deputies probable cause to believe Matheson's car contained contraband. This position finds support in several courts, including the United States Sixth Circuit Court of Appeals and the Georgia Court of Appeals. Those courts have held that a certification that a dog has been trained is prima facie proof of the dog's reliability which then may be rebutted by the presentation of evidence regarding the dog's performance or training. See United States v. Hill, 195 F.3d 258, 273 (6th Cir.1999); United States v. Diaz, 25 F.3d 392, 395 (6th Cir.1994); Warren v. State, 254 Ga.App. 52, 561 S.E.2d 190, 194-95 (2002); Dawson v. State, 238 Ga.App. 263, 518 S.E.2d 477, 481 (1999). "Although the dog's `credibility' may be undermined by evidence of its lack of training or past unreliability, the ultimate determination as to whether the dog is sufficiently reliable to support a determination of probable cause is for the trial court as the trier of fact." Dawson, 518 S.E.2d at 480.
When the evidence presented, whether testimony from the dog's trainer or records of the dog's training, establishes that the dog is generally certified as a drug detection dog, any other evidence, including the testimony of other experts, that may detract from the reliability of the dog's performance properly goes to the "credibility" of the dog. Lack of additional evidence, such as documentation of the exact course of training, similarly would affect the dog's reliability. As with the admissibility of evidence generally, the admissibility of evidence regarding a dog's training and reliability is committed to the trial court's sound discretion.
Diaz, 25 F.3d at 394.
"Prima facie" means that the proponent has fulfilled his duty to produce evidence and there is sufficient evidence for the court to consider the issue. Charles W. Ehrhardt, Florida Evidence § 301.2 (2002). Thus, the proposition advanced by the State is that the fact that a dog has been trained and certified to detect narcotics, standing alone, justifies an officer's reliance on the dog's alert to establish probable cause to search. But our review of the record and of pertinent literature convinces us that this is not enough.
Law enforcement use of narcotics detection dogs has become commonplace. And, generally, a trained dog's alert on a vehicle may constitute probable cause to search. See State v. Russell, 557 So.2d 666, 667 n. 1 (Fla. 2d DCA 1990); Denton v. State, 524 So.2d 495, 498 (Fla. 2d DCA 1988). The reason, of course, is the dog's keen sense of smell.
A dog's nose is uniquely equipped to detect the faintest of odors. Dogs possess potentially billions of chemical receptors called olfactory cells. These receptors are located among large supports inside the dog's nose named turbinate bones. Turbinate bones form numerous cylindrical passages that allow air exposure to millions more cells than is possible with simple tubular nasal passages, such as those found in human beings. Laid out, the surface area of these cells would cover a space the area of the skin on the dog's body.
In comparison, the surface area of human olfactory cells would cover no more than a postage stamp.
The effect of the dog's olfactory cells is not entirely clear. Some experts claim the result is an enhanced ability to detect minute levels of odorous material. Others assert that a canine's strength lies in its ability to discriminate among odors. Scientists supporting the discrimination theory believe that each olfactory receptor responds to a different odor; the more receptors, the greater the power to distinguish between scents. The answer most likely lies somewhere between the two opposing theories.
* * *
Little doubt exists that dogs have the ability to detect the smallest traces of odors and to perceive these scents much better than human beings.
Robert C. Bird, An Examination of the Training and Reliability of the Narcotics Detection Dog, 85 Ky. L.J. 405, 408-09 (1997).
Certainly, the olfactory superiority of dogs recommends their use by law enforcement. But, when determining probable cause to search, it can also be a weakness.
Many times the possibility of a false alert will be overlooked by a handler as will the dog's inability to differentiate between a "live" scent and a "dead" scent. Each dog will also vary in its ability to ignore detractors and masking agents[.]
Max A. Hansen, United States v. Solis: Have the Government's Supersniffers Come Down With a Case of Constitutional Nasal Congestion?, 13 San Diego L.Rev. 410, 416 (1976). Indeed, in this case Razor's trainer acknowledged the tendency of narcotics detection dogs to alert on the residual odors of drugs that are no longer present.
This underscores one of three central reasons why the fact that a dog has been trained, standing alone, is not enough to give an officer probable cause to search based on the dog's alert. Razor's trainer acknowledged that a trained dog, doing what he has been conditioned to do, imparts to the officer merely that he detects the odor of contraband. To be sure, as the trainer maintained, this may not be a false alert when assessing the success of the dog's conditioning. But for Fourth Amendment purposes it is neither false nor positive. The presence of a drug's odor at an intensity detectable by the dog, but not by the officer, does not mean that the drug itself is present. An officer who knows only that his dog is trained and certified, and who has no other information, at most can only suspect that a search based on the dog's alert will yield contraband. Of course, mere suspicion cannot justify a search. See Coney v. State, 820 So.2d 1012, 1014 (Fla. 2d DCA 2002). It follows that proof of facts that could justify only a suspicion cannot prima facie establish probable cause.
Another problem with predicating a finding of probable cause solely on the fact that a dog has been trained stems from inherent variables in the training endeavor. Although we commonly refer to the "training" of dogs, manifestly they are not trained in the sense that human beings may be trained. It is not a process of imparting knowledge and skills that dogs want or need. However much we dog lovers may tend to anthropomorphize their behavior, the fact is that dogs are not motivated to acquire skills that will assist them in their chosen profession of detecting contraband. Rather, dogs are "conditioned," that is, they are induced to respond in particular ways to particular stimuli. For law enforcement purposes, the ideal conditioning would yield a dog who always responds to specified stimuli in a consistent and recognizable way, yet never responds in that manner absent the stimuli. But this does not happen. While dogs are not motivated in ways that humans are, neither can they be calibrated to achieve mechanically consistent results.
As our record demonstrates, conditioning and certification programs vary widely in their methods, elements, and tolerances of failure. Consider, for example, the United States Customs Service regime:
The Customs Service puts its dog and handler teams through a rigorous twelve-week training course, where only half of the canines complete the training. Customs Service dogs are trained to disregard potential distractions such as food, harmless drugs, and residual scents. Agents present distractions during training, and reward the dogs when those diversions are ignored. The teams must complete a certification exam in which the dog and handler must detect marijuana, hashish, heroin, and cocaine in a variety of environments. This exam and the following annual recertifications must be completed perfectly, with no false alerts and no missed drugs. If a dog and handler team erroneously alerts, the team must undergo remedial training. If the team fails again, the team is disbanded, and the dog is permanently relieved from duty.
Bird, 85 Ky. L.J. at 410-11. In contrast, the testimony below disclosed that Razor and his handler had undergone just one initial thirty-day training course and one week-long annual recertification course. In neither course was Razor conditioned to refrain from alerting to residual odors. Whereas the Customs Service will certify only dogs who achieve and maintain a perfect record, Razor's certification program accepted a seventy percent proficiency. These disparities demonstrate that simply characterizing a dog as "trained" and "certified" imparts scant information about what the dog has been conditioned to do or not to do, or how successfully.
Finally, dogs themselves vary in their abilities to accept, retain, or abide by their conditioning in widely varying environments and circumstances. "[E]ach dog's performance is affected differently by working conditions and its respective attention span. There is also the possibility that the handler may unintentionally or otherwise prompt his dog to alert." Hansen, 13 San Diego L.Rev. at 416. The Customs Service monitors its dogs' performance in the field. Recognizing that a dog's ability can change over time, it maintains records for only thirty to sixty days, then discards them because older records are not probative of the dog's skills. Bird, 85 Ky. L.J. at 415. The Hillsborough County Sheriff's Office maintained no records of Razor's performance, and his handler had not kept track.
For these reasons, we conclude that the fact that a dog has been trained and certified, standing alone, is insufficient to give officers probable cause to search based on the dog's alert. One Florida case has recited additional factors that must be known in order to conclude that an alert by a narcotics detection dog is sufficiently "reliable" to furnish probable cause to search. In State v. Foster, 390 So.2d 469 (Fla. 3d DCA 1980), the Third District identified these factors as
the exact training the detector dog has received; the standards or criteria employed in selecting dogs for marijuana detection training; the standards the dog was required to meet to successfully complete his training program; the "track record" of the dog up until the search (emphasis must be placed on the amount of false alerts or mistakes the dog has furnished).
Foster, 390 So.2d at 470 (quoting Hansen, 13 San Diego L.Rev. at 417). We agree with this list of factors, and we especially join in the Foster court's emphasis on the dog's performance history. A dog's alert can give an officer probable cause to search only if the officer reasonably believes that the dog would not exhibit the alert behavior unless contraband was present. Given the "language barrier" between humans and caninesthus, for example, preventing the officer from questioning the dog further for corroborative details, as he might a human informant the most telling indicator of what the dog's behavior means is the dog's past performance in the field. Here, the State did not present any evidence of Razor's track record. Accordingly, we conclude that the State did not meet its burden to establish that the deputies had probable cause to search Matheson's car.
We note that, even if we were to accept the State's position that it made a prima facie showing of probable cause founded solely on the fact that Razor was trained and certified, that showing was rebutted as a matter of law. The deputies' own undisputed testimony at the suppression hearing established that they knew that Razor's reliability for detecting the presence of contraband in the field was ungauged and that it could not be predicted based on his particular conditioning. In light of these facts, Razor's alert could not have given the deputies probable cause to search under any test.
We reverse the denial of Matheson's motion to suppress and remand with directions to discharge him.
FULMER and STRINGER, JJ., Concur.