A dog’s nose knows: Canines detect traces of gasoline down to one billionth of a teaspoon
The uncanny ability of dogs to smell minuscule amounts of gasoline or other accelerants is a boon to arson investigators.
Unfortunately, Sparky isn’t the best expert when it comes to a court of law.
As part of a project to build better laboratory tests to confirm a dog’s findings, a team of University of Alberta researchers set out on a mission to determine how sensitive a dog’s nose really is — and they learned it is really, really sensitive.
“The idea here was to come up with a new protocol that allows us to do comprehensive testing of the dogs and identify that number,” said Robin Abel, a graduate student who worked about six years with the RCMP before pursuing a chemistry PhD.
“We found that the dogs were far more sensitive than previously estimated. It looks like they’re capable of detecting somewhere around one-billionth of a teaspoon of gasoline.”
It’s not easy to describe just how small that is but one way might be to imagine the size of a drop of ink from an inkjet printer — then think of something much smaller than that, said James Harynuk, a chemistry professor and Abel’s supervisor.
Laboratory corroboration of a dog’s evidence is important to protect against false positives and wrongful convictions, as well as to provide greater context for investigators and the courts.
But current lab methods aren’t nearly as sensitive as a dog’s nose in tracing ignitable liquids. Abel is in the process of developing lab tests that are more sensitive — but to do that, he needed a way to figure out the absolute smallest amount that a dog could detect, the researchers said Tuesday in an interview with CBC News.
“Right now, there’s a question of how many validations is the dog getting wrong, how many people might we be convicting who haven’t actually done anything wrong, because such a very, very small amount of ignitable liquid has been found,” Abel said.
“Being able to match up the lab to the dog’s sensitivity allows us to have more information about that, which would allow us to better determine whether or not yes, this person has something, this is a significant finding or no, the dog responded to something that makes sense but doesn’t mean anything in terms of forensics or law.”
As fascinating as it was to learn more about canine sniffing sensitivity, the bigger point of the research, published this month in Science Direct, was to come up with a protocol for producing extremely clean, porous tile material that can be used for evaluating dogs and their detecting abilities, Harynuk said.
The researchers believe that technology could be useful in other situations where dogs need to be trained or focused on very particular smells.
Despite their own research achievements, Abel and Harynuk agree that the naturally ability of a dog to smell and process information sets a pretty high bar for scientists to match.
“It only takes the length of a sniff,” Abel said. “We have to prepare the sample and do a full analysis and then look at the data and then we can say one way or another. They’re just so much quicker than we can be. And at a fire scene, when it’s just piled with ash and debris, that is a huge advantage.”