From rescue to research: training detection dogs for conservation
- Conservation and research teams have used detection dogs to locate illegal wildlife products, weapons, invasive species, and, particularly, wildlife scat–a non-invasive way to collect dietary, hormonal, and genetic information contained in fecal material.
- Training detection dogs builds on their obsessive drive to play by associating a target substance with the play reward.
- Handlers are instrumental in interpreting a dog’s behavior and ensuring it searches efficiently and effectively for its targets.
- Detection dogs are a cost-effective way to collect wildlife data, though the costs of international transport may limit their use by smaller conservation groups.
Ecologists on Santa Cruz island off California had a big problem with small creatures. Argentine ants, an invasive species that disturb native bees, ants, and other insects critical to the island’s ecosystem, are nearly impossible to eliminate.
Nature Conservancy reserve managers were eliminating the ants through aerial delivery of poisoned sugar water droplets, but to be sure no colonies of the invader remain, they hired a dog.
People have long used dogs’ amazing sense of smell to search for drugs and explosives, lost people, rare species, and even human diseases. Research and conservation teams use dogs to find poachers, illegal wildlife products, and wildlife dung—or scat, often for multiple species, a non-invasive way to collect dietary, hormonal, and genetic information contained in the fecal material.
Tobias the detection dog knows the smell of Argentine ants better than any human and can locate them even over wild terrain. He’s an example of a “technology” that helps field teams across the world find snares, intruders, and wildlife.
From drugs to dung: detection dogs trained for conservation
But how do Tobias and his canine (and human) companions learn their trade? Who trains these dogs, and how?
Three groups—Working Dogs for Conservation (WD4C), Conservation Canines, and PackLeader—have trained many of the dogs used by projects around the world to detect wildlife and related substances, including scat, carcasses, and processed products, such as rhino horn or ivory. All started in the mid- to late-1990s adapting training from police and narcotics-detection dogs.
Now, WD4C partners with government agencies, protected areas, non-profits, and researchers at project sites on five continents. The group’s conservation biologists also help teams to design studies, collect data, create GIS maps, analyze data, and report on the data collected.
Conservation Canines, at the University of Washington Center for Conservation Biology, uses scat-detecting dogs in combination with lab analysis to extract a variety of genetic and physiological indicators from the scat. The initiative began in 1997 when Dr. Samuel Wasser, the Center’s Director, teamed up with a police dog trainer to teach narcotics detection dogs the new tricks of detecting scat from endangered species. Since then, the team has used dogs to monitor the presence of a wide range of wild terrestrial, aerial, and aquatic animals (typically their scat).
Why are dogs so good at detecting specific substances?
Dogs are built to smell. They have nearly 50 times more scent receptors in their noses than we have, plus a 40-times larger portion of their brain dedicated to smelling. It’s no wonder scientists believe dogs’ sense of smell is more than 10,000 times as acute as ours. They can, for example, detect a human smell from far away and can distinguish one person from another by smell alone.
Dogs can smell continuously (not just when they inhale, as we do), and they can tell which nostril receives an odor first and thus the general direction of where the scent is coming from, a huge plus in searching in unknown environments. Searching by scent allows conservation detection dogs to detect plants, animals, and scat that people can’t, especially those hidden by terrain, vegetation, or water. Moreover, they can locate multiple targets, such as scat from multiple species, and can do so in a forest or grassland more quickly than humans can using our eyes.
“Bad” dogs can become good detectives
But a successful working dog must have more than just a good nose.
Scent detection dogs are typically focused, high-energy animals obsessed with a favorite toy, a combination that can frustrate owners but keeps the working dogs eager. In fact, all three groups find many of their dogs at animal shelters and include in their mission the rescue of these dogs, who which are hard to place in homes and thus, in many cases, might otherwise be euthanized.
Conservation Canines works with shelters to find energetic ball-fanatical dogs, typically those that detect airborne scents, rather than those that follow a track with their nose close to the ground.
Many dogs are able to detect target objects, yet WD4C screens up to 1,000 dogs for each one they accept. It’s that obsessive toy drive that motivates certain dogs to work so hard to find target scents that makes them successful. Ideal detection dogs are persistent—“dogged,” if you will. They have, as WD4C co-founder Megan Parker, says in a TED-x talk, an “inability to quit.”
Trainers also look for dogs with ’nerve strength’—the capacity to work efficiently in unfamiliar environments and conditions. “[It] means they can handle all the crazy stuff that happens around them,” said Pete Coppolillo, Executive Director of WD4C. “We move dogs around in trucks and boats and ATVs and airplanes and even once on elephant back, so they need to stay cool during all that.”
What goes into detection dog training?
Nevertheless, said Coppolillo, “Dogs first have to learn to be a detection dog. In other words, they need to understand the ‘game’ they’re playing— and it is a fun game for them.”
Training, which for WD4C can take three to four months, matches the finding of a target scent with a reward—the chance to play with a ball or other toy—that the dogs crave. Each time a dog locates a new sample, it gets a chance to play with its ball or other toy.
WD4C first pairs the dog’s toy reward with a target scent–for example, hiding their favorite ball with bushmeat. Gradually, the trainers separate the scent from the toy, so that when a trained dog finds the target scent, it knows to sit and look at its handler, who delivers the toy.
“Once they’ve got that process down,” said Coppolillo, “it’s pretty straightforward to introduce new scents into the same game. Veteran dogs learn new scents pretty quickly because they know that it’s the same game, but with a different target scent.”
Conservation Canines can teach an experienced dog the scent of a new species in a day but “to solidify their ability to locate these [generally rare] species, we take a number of weeks to acclimate them to the scent.” Once trained, a dog can learn new targets, include the scats of more than 10 different species.
Trainers ensure that dogs find targets regularly enough (and thus receives plenty of play rewards) to keep them motivated and make efficient use of their and their handler’s time in the field. If a target is rare, such as Argentine ants on Santa Cruz island, trainers can hide additional samples, add a new species, or visit sites with more of the target substance to keep dogs motivated.
Dogs are trained to stick with the smell of the target substance without getting distracted by other animals or their scat. A professional detection dog should not show interest in the actual animals associated with the target scat—they are meant to be non-invasive and ignore wildlife while in the field.
However, dogs trained to detect scat of several species will look for all of these at the same time, so another management issue is the training of dogs on the scat of sympatric species (those living in the same area), some of which are more common than others.
“For example,” Coppolillo said, “a dog trained on white-tailed deer scat, would be stopping every 50 feet throughout most of the U.S., because deer are so abundant and widespread. That would make it very difficult for that dog to range farther to find bear or lynx or wolverine scats, which occur at much lower densities.”
Similarly, if the Conservation Canines team thinks a sympatric species the dog is trained on will cause difficulties for a project, they will prepare a new dog that is not already trained on that species.
Where dogs can help
For what types of project might field teams use a detection dog (or two)?
Conservation groups have most commonly used detection dogs to detect and monitor rare species, above and below ground, and underwater, including whales and invasive fish.
Trainers can teach dogs to sniff out particular substances of interest to a partner, and these have changed over time. “Five years ago, most of our work was on carnivore scats,” said Coppolillo, “but now, in addition to the carnivore scat, we work on exotic weeds, aquatic invasives, insects, all the trafficked wildlife targets like ivory and rhino horn, and guns and ammunition. We’ve even got dogs sniffing illegal hardwoods now, so we’re all over the map as far as scents go.”
He highlighted the value of using dogs to detect rare targets, such as the last remaining Argentine ants. “Dogs are most useful and cost effective when the targets they’re looking for are rare,” he said. “I call it the ‘first individual/last individual’ principle—when you’re looking for the first invasive plant, or trying to get rid of every last invasive beetle (or their eggs or larvae), dogs are a great way to go.”
Who accompanies a detection dog on the job?
All the training centers match each dog with a particular human, the handler, for a specific project. In fact, PackLeader stresses, “Selection of a handler can often be more critical than the selection of a dog.”
Training the handler improves the dog’s efficiency and effectiveness in locating targets. Handlers must learn to correctly identify scat and avoid showing interest in scat of other species, which could encourage a dog to look for incorrect scat in order to please the handler and get a reward.
“Someone who is handling a detection dog has the primary job of watching the dog at all times,” explained Karen DeMatteo, a researcher working with PackLeader to help identify conservation corridors in Argentina, in an email to Mongabay-Wildtech. “[Having] a handler that can catch the changes in behavior that are more subtle can help the dog locate those samples that might be located a little farther, a little older, or more hidden.”
During training, the handler learns to read and interpret the dog’s movements and any changes in behavior. Training novice handlers can take several weeks of hands-on work to learn how to select, take care of, understand, and motivate a detection dog. They also learn about smells—their characteristics and how dogs detect them, effects of the environment on the scents of target objects, and how to make best use of the dog’s abilities to searching for target objects.
For these reasons, Conservation Canines does not rent out its dogs without either training on-site handlers or having its own experienced handlers accompany the dogs in the field.
Costs and other considerations
Carly Vynne, a conservation biologist who worked in the field with detection dogs for her doctoral research at University of Washington, contributed a Wildtech story about how these dogs have contributed to research and conservation. She recommended that wildlife professionals consider several long-term commitments before investing in detection dogs.
Key considerations, she explained, “include the fairly large start-up costs, as well as upkeep: dogs require housing, maintenance, specialized care (in our case, an hour of de-ticking each afternoon), training, and people that specialize in working them in the field. This is an investment that doesn’t start or stop at the life of a given field project, so leasing dogs is often a good option, and several programs offer this service.”
Both Vynne and other researchers found detection dogs to be cost-effective in surveying wildlife. Researchers in Vermont found that detection dogs more effectively located bear, fisher, and bobcat scat than did other methods. “Although the cost of surveying with detection dogs was higher than that of remote cameras or hair snares,” they wrote, “the efficiency of this method rendered it the most cost-effective survey method.”
Despite the effectiveness of dog-directed scat searches, the cost of paying U.S. handler salaries and contract fees for dog-handler teams (US$1,479 per sample in a study of Cross River Gorillas in Cameroon) may still be prohibitive for many field projects.
Researchers working in Cameroon and China have recommended establishing regional detection dog training programs that could serve various field projects. Using local dogs (less susceptible to disease) and handlers would enable many more field teams to use dog “technology” cost-effectively without having to hire and transport dogs internationally.
The researchers in China, who used a detection dog to find scat from several primate species, also recommended collaborating with local police to train the dogs locally, which can reduce costs and bureaucratic obstacles while building local capacity to enable long-term studies.
Recommendations for groups thinking of employing detection dogs
In a 2016 review of detection dogs for conservation, Beebe and colleagues concluded, “A lack of available information regarding the selection process is likely to make it extremely difficult for conservation groups to identify suitable dogs, upon which scarce resources can then be expended for training and deployment purposes.” The authors urge conservation groups considering investing in a detection dog to first consult with experts about the various characteristics most likely to predict success.
Groups can consult the Conservation Canines FAQ page, which provides quick answers to a number of questions on their program.
Coppolillo also recommends that field teams consult with a reputable trainer to ensure realistic expectations and a positive result. He suggested specifying what your team wants from using detection dogs and what the cost is of being wrong.
“If you’re spending $140 per sample to get fisher DNA from scats,” said Coppolillo, “a dog finding a whole bunch of pine marten scats instead of fisher will break the budget! If you really need good samples and/or reliable data, it’s worth spending the money for an experienced dog-handler team, preferably one that has already worked on the target species.”
|A fisher with its winter coat, for which it was nearly wiped out two centuries ago. Photo credit: ForestWander CC 3.0||The closely related but smaller pine marten, was also prized for its fur. Photo credit: Cody Connor CC 2.5|
Arandjelovic, M., Bergl, R. A., Ikfuingei, R., Jameson, C., Parker, M., & Vigilant, L. (2015). Detection dog efficacy for collecting faecal samples from the critically endangered Cross River gorilla (Gorilla gorilla diehli) for genetic censusing. Royal Society open science, 2(2), 140423. Chicago http://rsos.royalsocietypublishing.org/content/2/2/140423?utm_source=TrendMD&utm_medium=cpc&utm_campaign=Royal_Soc_Open_Sci_TrendMD_0
Beebe, S. C., Howell, T. J., & Bennett, P. C. (2016). Using Scent Detection Dogs in Conservation Settings: A Review of Scientific Literature Regarding Their Selection. Frontiers in veterinary science, 3. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5083854/
Long, R. A., Donovan, T. M., Mackay, P., Zielinski, W. J., & Buzas, J. S. (2007). Comparing scat detection dogs, cameras, and hair snares for surveying carnivores. Journal of wildlife Management, 71(6), 2018-2025. http://www.bioone.org/doi/abs/10.2193/2006-292
Orkin, J. D., Yang, Y., Yang, C., Yu, D. W., & Jiang, X. (2016). Cost-effective scat-detection dogs: unleashing a powerful new tool for international mammalian conservation biology. Scientific reports, 6. https://www.nature.com/articles/srep34758
Vynne, C., Skalski, J. R., Machado, R. B., Groom, M. J., Jácomo, A. T., Marinho‐Filho, J.A.D.E.R., … & Wasser, S. K. (2011). Effectiveness of Scat‐Detection Dogs in Determining Species Presence in a Tropical Savanna Landscape. Conservation Biology, 25(1), 154-162. http://onlinelibrary.wiley.com/doi/10.1111/j.1523-1739.2010.01581.x/full
Wasser, S. K., Davenport, B., Ramage, E. R., Hunt, K. E., Parker, M., Clarke, C., & Stenhouse, G. (2004). Scat detection dogs in wildlife research and management: application to grizzly and black bears in the Yellowhead Ecosystem, Alberta, Canada. Canadian Journal of Zoology, 82(3), 475-492. http://www.nrcresearchpress.com/doi/abs/10.1139/z04-020#.WiBuMlXtzDc