Article questioning police scenting dog reliability - Page 1

Time magazine article on "Is dog-scent evidence junk science?"

http://www.time.com/time/nation/article/0,8599,1913987,00.html?cnn=yes

This is based on the dog handler who has been proven to be a fraud.

From the article:
"In detective dramas, a dog's powerful sense of smell has become a predictable crime solver: the trusty canine takes a sniff of a suspect object and follows the scent, eventually coming to the perpetrator of the evil deed. But in real life, is this reliable evidence — or is it junk science that has helped put away innocent people?"

Similar Florida Today article here:

http://beta.floridatoday.com/article/20090621/NEWS01/906210319/-1/sevendays

I haven't read the articles yet, but I will.  I can tell you that scent discrimination tracking works from first hand experience. 

I recently tracked a suicidal subject that ran off into the woods at a park.  An astute officer had the subjects family bring a shirt of his before I arrived.  When I got on scene the subject had been missing for 45 minutes.  I had my dog take scent from the shirt and the dog immediately started tracking.  The first officer on scene was directing me to the last place the subject was seen, the dog was already tracking that way.  The cop kept saying "ok, go right up here...oh your dog is going there, go left at that trail...oh, the dog is there."  We tracked through very heavy vegetation then out to a street, about 3/4's of a mile through a neighborhood on the street and back into a wooded area where we found the subject.  The whole track was easily over a mile, included yards, streets, thick woods and plenty of people out contaminating the path.  Was a hot day and poured rain 1/2 way through the track.  I asked the guy where he went, just to see if the dog was on him the whole way?  The guy described our exact track including yards and streets.  So, I know it works.  We passed plenty of other people and the dog showed no interest in them. 

We train scent discrimination tracking a lot for our patrol dogs, in addition to training ground disturbance and air scenting.  Bernhard Flinks changed the way our K9 unit tracks, he is incredible.  He has done two seminars for me and my PD and they were awesome.  Depending on the situation and type of call it gives me a whole bunch of options.  We recently had a high profile murder case in our city, the case was basically solved by one of our K9's tracking, finding evidence including the murder weapon and tracking to the guys door.  The DNA recovered along with the fact that the dog tracked to his door sealed the conviction. 

The biggest problem we have in tracking in an urban environment is that most bad guys have get away cars waiting close by, or cell phones to call for rides.   We get lucky when their plan goes to shit and they can't get to their car and run on foot, that's when we can track them and catch them.  It's a great feeling when that happens.

Jim

There have been some frauds, in the past. One gal had a Dobermann Pinscher with flop ears, as I recall. She was doing volunteer work for the local police. She was a self-made expert. I cannot remember her name, but she finally was outed and disgraced.. not before she allegedly put some innocent people in jail and led police on some wild goose chases. Some people are nut jobs and like to hang out with cops for the sense of self-importance and attention it gets them. These people should be avoided, at all costs.

I read the articles, very interesting.  I would never make the claims made in the article about my dog's abilities.  Science doesn't even truly understand the dogs olfactory system and what the dog can detect.  We can make machines to do just about everything including space travel and unmanned aircraft over the battlefield but we haven't built one yet that works as well as a dogs nose. 

The art is reading the dog and knowing what you are reading form his behavior and body language.  I don't think we have perfected the system well enough for "scent lineups" here yet.  It's nice having fingerprints and DNA for the science part and the dog to lead you to the evidence.  Then you have a solid case

JMO,

Jim 

"We can make machines to do just about everything including space travel and unmanned aircraft over the battlefield but we haven't built one yet that works as well as a dogs nose. "

But it's certainly in the works.  Here's a description of an ongoing program by DARPA.

RealNose

Despite the recent development of a wide variety of chemo-sensors, the current technologies for accurate detection and identification of known and unknown chemicals and mixtures of chemicals (i.e., explosives, chemical and biological weapons) possess limited performance range with regard to odorant identification, odorant recognition in mixtures, and chemo-sensing across multiple classes of compounds. The current detection systems also stand in stark contrast to the innate ability of the mammalian olfactory system―specifically the canine system―to handle these challenging tasks with ease. A detection system that more closely models the canine olfactory system offers greater sensitivity to a wide range of new and old chemical agents, and will serve to protect troops and infrastructure.

The goal of the DARPA RealNose Program is to build a "nose" constructed from actual olfactory receptors that further leverages the components of the canine olfactory system to create a breakthrough detection system with potential capabilities beyond that of a canine.

The key to the program concept is that by simulating the entire mammalian olfactory system (from air intake to pattern recognition), revolutionary detection capabilities will be created, demonstrating canine-comparable specificity, distance, and detection thresholds. Approaches currently under investigation include novel olfactory receptor expression and stabilization techniques, signal transduction procedures, and revolutionary pattern recognition methodology.

The technologies developed in this program will greatly increase our detection cability to recognize and identify a wide range of current and emerging chemical agents.

That would be great if they can do it.  I would also need to be portable, cost effective and have the ability to check above, below and in items.  I can see the technology being developed for exposives and chemical agents first.  It would then be easy to "calibrate" it for narcotics, similar to the ion scan machines at airports, that technology can also do dope if calibrated for it.

I think it's still a long way off.  Amazing what the future will bring.

Jim