Man's best friend proves useful in detecting threatened wildlife

Fort Irwin National Training Center is looking to expand its facility by some 120,000 acres and at the same time protect threatened species there.

Dr. Mary Cablk, DRI assistant research professor, is assisting the U.S. Army at Fort Irwin, near Barstow, Calif., in this effort by collaborating with tortoise biologists from the U.S. Geological Survey's Biological Resources Division, academia and the private sector.

They are colleagues in a required management and research program that will include locating, "translocating"—or displacing—researching and ultimately protecting the desert tortoise, a federally designated threatened species native to the Mojave Desert.

With such a large expanse of land to survey, it would take humans an incredible amount of time to track and remove the thousands of tortoises estimated to need translocation to other protected areas nearby.

Given that people have been shown to have limited success rates at finding tortoises—only 30 to 60 percent for two size/age classes of tortoises—Cablk and her colleagues are moving forward with a program that has, in its preliminary stages, proven considerably more effective than using humans alone.

Cablk has teamed up with man's best friend to develop another viable option: let the four-legged species with a keener sense of smell find the tortoises instead.

All in a day's work: Left to right, dog handler Aimee Hurt and "Fin," researchers Jill S. Heaton and Mary Cablk, dog handler Alice Whitelaw and "Camas" pose for a group shot.

Cablk is no stranger to understanding the abilities of search-and-rescue dogs, as she and her own German shepherd are auxiliary deputies who specialize in finding missing people and crime victims.

If dogs can find people or drugs or bombs, why not use them to find tortoises?

Using her professional expertise in remote sensing technology and landscape ecology, Cablk joined forces with the Redlands Institute, of the University of Redlands (Calif.); the U.S. Army Research Office; the U.S. Fish and Wildlife Service; and the Bureau of Land Management Desert Tortoise Conservation Center to determine if certain wildlife detection dogs could be trained and relied upon to locate and correctly identify desert tortoises under semi-natural conditions.

"The more tortoises we find, the more data we will collect about the population. Therefore, we can give better estimates about population numbers and structure," Cablk says about why improved survey tools are being explored.

"This is a once-in-history opportunity to make significant advances in our understanding about the desert tortoise in a very short time based on massive amounts of new data. The results, we hope, will reverse the negative trend in tortoise populations."

The Players

Detection Dogs: Cablk brought in dog teams from the non-profit organization Working Dogs for Conservation that specialize in scat detection of rare, threatened and endangered mammals. Dog handlers worked with the dogs in their native Montana to introduce the tortoise scent and train the dogs to understand there was a reward for targeting that scent. The transition to live tortoises was done on site at the Desert Tortoise Conservation Center outside of Las Vegas in the Mojave Desert where tortoises live. It takes a special dog with the right type and level of drive to do live animal detection work off leash. The dogs need to be motivated to work and when they reach a tortoise, inherently want a reward from their handler, not from the tortoise itself. Not everyone is the right type of handler, either. Tortoise detection work requires both the dogs and the handlers to be athletes and to be able to communicate to each other consistently, clearly and effectively.

Tortoises: After obtaining the required permit from U.S. Fish and Wildlife Service, Cablk and colleague Dr. Jill Heaton from the University of Nevada, Reno were able to use research tortoises at the Desert Tortoise Conservation Center for the study. In each trial, known numbers of tortoises were tethered at spatially random locations to prevent them from relocating more than a half-meter in any direction during the trial. The tether design was also approved by U.S. Fish and Wildlife Services to ensure minimal stress to the tortoises from heat, cold, restraint and potential predation. Hundreds of dog-tortoise interactions were carefully logged using an extensive relational database designed to record and maintain spatial and statistical data.

Handler Alice Whitelaw and German shepherd "Camas" demonstrate that dogs can find tortoises safely and reliably. Camas has picked up the scent of the tortoise in the air and is about to indicate this fact to her handler by doing her trained alert freezing in place and staring at Whitelaw.

Humans: Cablk's research team included hand-picked specialists, each of whom contributed unique but critical expertise that would maximize success of the pilot project. Cablk's understanding of scent detection and tortoise issues enabled the team to communicate across their respective areas of expertise. First, a master dog trainer from Contra Costa County, Calif., was selected for her skills in canine behavior and specialized technical-detection training. Heaton, a tortoise biologist, oversaw the biological element related to tortoises and their habitat. She brought with her a project manager skilled in database design to ensure continuity of the complex spatial and temporal data set collected during the one-month study. Add the dog handlers, three database GIS/GPS technicians and more than 10 student workers to round out the team. And of course, the study designer, Cablk: "I speak the dog language and science language, which was the glue to make sure we met our objectives."

The Study

The dogs and their handlers were fitted with GPS units to record their exact locations in one-second intervals. Each time a dog identified a tortoise, the event was recorded. During the trials, climate conditions at dog-nose height were recorded using specially designed meteorological stations to determine if there were identifiable trends in environmental conditions, such as wind speed, wind direction, air temperature and relative humidity that correlated with detection distances.

The Results

With an overall accuracy of more than 90 percent for finding tortoises on the surface and in burrows, the dogs were recorded to locate tortoises at distances of more than 60 meters, about 200 feet. Their rate of detection was unaffected by the size or sex of the tortoise or the time of day.

Climate conditions were not a factor in detection distances-a finding that contradicts conventional wisdom about what comprises "ideal" scent cond itions. The dogs were able to locate very small tortoises that humans do not survey due to difficulty in finding and even detected extra tortoises not involved in the study.

Factors to Consider

Tortoise and dog safety was the primary concern in the trials. During the course of the training, three dogs were removed from the program due to safety concerns resulting from excessive "prey drive." Two exhibited aggressive behavior toward the tortoises and one was not experienced enough to do the intense field trials.

These dogs lacked the acceptable hunt and play drive respectively for wildlife detection work. Natural hazards existed but were uncontrolled, such as rattlesnakes, cacti, weather changes and rough terrain. The dogs and their handlers took these hazards in stride as would be expected under actual working conditions.

The Next Step

Now that the trials are completed, Cablk is hard at work analyzing the data to create a program outlining the boundaries for using wildlife detection dogs to locate threatened or endangered species, the success rate anticipated, how many people it would take to execute a job and the length of time needed to get the job done satisfactorily. She also is compiling results to determine what constitutes a good dog team.

"Would you want just any dog checking an airplane you were getting on instead of a trained bomb dog?" Cablk asks. "No, and that's the type of rigor we brought to researching appropriate wildlife detection dog teams."

She will continue research into the dogs' drive and the level of training required, including obedience and strategic search aspects, as well as narrowing the dog-screening process by looking at the history of their training.

Cablk is grateful to the University of Redlands for its year-long development of an intricate database design that tracks everything from where the tortoises were located, their measurements, sex and age, to which handler was with which dog and the weather conditions during the trials.

The glass ceiling for dog career choices may have just come crashing down with Cablk's research results. No longer are dogs limited to law enforcement, medical screening, pest detection and search-and-rescue. The career field seems to be expanding to include conservation efforts for fellow animals.

-Heather Emmons