Does normal paper currency contain enough narcotics residue to attract a drug-sniffing dog?

Question

It seems pretty well established that paper currency (e.g. in the US) is commonly contaminated with trace amounts of cocaine or other illegal narcotics. (See Snopes for instance, though I'm happy to have this assumption challenged.) Is this contamination sufficient to attract alerts by drug detection dogs?

Wikipedia says yes: "The drug content is too low for prosecution but not too low to trigger response to drug-sniffing dogs". But their citation is to a newspaper article that makes no mention of money or currency.

This 1998 Slate article says: "In 1994, a U.S. Circuit Court held that ordinary money contains enough cocaine to attract a drug-sniffing dog." They don't give a reference to the case, but I think it might be US v. Florez; however, the court's opinion only seems to say that the defense claimed this was true (note 12), and doesn't seem to come to a conclusion as to whether it actually is. There is also US v $30,060 in US Currency, in which the court found that a dog's alert on currency is not evidence that the possessor is somehow involved with narcotics, partly due to the fact that money is commonly contaminated. But this seems like a somewhat different question and a different standard of proof, and anyway I'm not inclined to consider judges as scientific experts.

On the other hand, the American Society of Canine Trainers says that "currency in circulation does not contain enough narcotic scent for a narcotic detector dog to alert to", and they also cite case law in their favor.

What data actually supports any of these claims? Does currency commonly contain enough narcotics residue to trigger an alert by a typical drug detection dog, at a rate significantly higher than the usual rate of false positives for such dogs?

[I became interested in this claim after it appeared in this travel.SE answer. Thanks RoboKaren, even though you're not a notable source by yourself.]

Answer 1

Evidence:

1. Research has shown that drug detection dogs act routinely based on the behavioral cues of their handlers, rather than only acting on their sense of smell for odor detection.

   In conclusion, these findings confirm that handler beliefs affect working dog outcomes, and human indication of scent location affects distribution of alerts more than dog interest in a particular location. These findings emphasize the importance of understanding both human and human–dog social cognitive factors in applied situations. Source: Handler beliefs affect scent detection dog outcomes

2. Research has also showed that some of the high odor compounds are not used in the manufacture of training scents used in training drug detection dogs which might lead to failure of detection of those drugs.

   A small number of volatile and semi-volatile compounds present in very low concentrations and associated with very low odor detection thresholds cause the characteristic smell of a drug. These high odor impact compounds are not being used to manufacture surrogate training scents used in training forensic canines. This omission could explain why these surrogate scents are generally not effective. This information could lead to increased understanding of what drug detection canines are using as the signature odor of street drugs. Source: Investigating the aroma of marijuana, cocaine, and heroin for forensic applications using simultaneous multidimensional gas chromatography - mass spectrometry - olfactometry

3. Various factors such as breed type, drug type and searching environment type might influence drug detection performance in drug detector dogs.

   The olfactory acuity of dogs’ sense of smell toward various volatile chemical compounds may differ considerably, though results may also reflect different experimental designs of different laboratories. Odors of different drugs may be differently sensed by dogs, and consequently ease of detection may differ. These differences may be related to polymorphic forms of olfactory receptor genes or their breed specific allelic variants or to the proportion of functional vs. non-functional genes showing affinity to the volatile chemical compounds characteristic of a drug. It is well known that scent detection dog performance depends not only on olfactory acuity but also on canine cognitive and learning abilities. The detection performance of sniffer dogs is context-dependent. Source: Efficacy of drug detection by fully-trained police dogs varies by breed, training level, type of drug and search environment

TL;DR:

Based on current research, microgram levels of cocaine present on circulated US currency is insufficient to draw an alert from drug detector dogs.

The authors concluded that as the average level of cocaine present in a single bill (10 lg) is 100,000 less than the average level required for drug-detector canine alert (1 g), “it is not plausible that innocently-contaminated US currency contains sufficiently enough quantities [sic] of cocaine and associated volatile chemicals to signal an alert from a properly-train drug detector dog.”. Source: Drug Contamination of U.S. Paper Currency and Forensic Relevance of Canine Alert to Paper Currency: A Critical Review of the Scientific Literature.

However there are only two published experimental studies on drug-detector dog alerts to U.S. currency spiked with various amounts of cocaine for the above comment and more research is needed to drawn an conclusion whether normal US paper currency contains enough narcotics residue to attract a drug-sniffing dog.