Wildlife Detection Dogs Take Lead in Locating Threatened Desert Tortoises


The following contains a summary of information regarding the U.S. Army Research Office and University of Redlands funded project using K9s to survey missing cohort groups of Gopherus agassizii (Mojave Desert Tortoise).

Threatened Species

In 1990, the U.S. Fish and Wildlife Service listed the Mojave population of the desert tortoise as a threatened species. Since then, federal, state, and local wildlife and land management agencies and jurisdictions have become actively involved in programs to promote conservation and recovery of the Southwest’s most beloved animal. A forward-looking project, led by Dr. Mary Cablk, Desert Research Institute (DRI), and Dr. Jill Heaton, University of Nevada Reno, has shown that wildlife detection dogs may provide a real advantage in surveying populations of this threatened species. Accurate information about tortoise populations, including breakdowns into age classes, such as adult, sub-adult, juvenile, and neonate, is critical in developing effective conservation strategies to reverse what is believed to be a decline in populations. In their pilot project, trained detector dogs were studied to evaluate their usefulness in sniffing out tortoises in their natural desert habitat. This study was funded by the Redlands Institute at the University of Redlands in California with cooperation from partnering agencies, including the U.S. Fish and Wildlife Service and the U.S. Bureau of Land Management’s Desert Tortoise Conservation Center (DTCC) in Las Vegas, Nevada.

Increasing concern over declining populations of rare, threatened, and endangered species is spurring the development of noninvasive methods for censusing, monitoring, and studying these species. Noninvasive approaches to wildlife surveys often rely on “sign” (such as scat [feces], urine, hair, or dens) to detect and track animals without imposing significant disruption on already imperiled species. Finding the species of interest—or target species—can be difficult depending on the size, range, and habits of the species. Well experienced in understanding the abilities of detection dogs, Dr. Mary Cablk believed dogs could fit into the solution for recovering the tortoise populations of the Mojave Desert.

Researchers, dog handlers, and dogs at DTCC including (left to right) WDC dog handler Aimee Hurt with Fin; Dr. Jill Heaton, UNR; Dr. Mary Cablk, DRI; and WDC dog handler Alice Whitelaw with Camas.
K9DT Team
click to enlarge and view the team members
Expanding the body of reliable knowledge about a threatened or endangered species is crucial in improving chances to protect and preserve the species. From this perspective, the study designer, Cablk, an expert in remote sensing technologies and landscape ecology, carefully assembled a team of top-notch specialists in this pilot project to collect critical data about desert tortoises in a scientifically rigorous manner. The team included Dr. Heaton, a desert tortoise biologist; a master dog trainer; three GIS/GPS (geographic information system/global positioning system) database technicians; a cadre of student workers; and professional dog handlers from Working Dogs for Conservation (WDC), a nonprofit organization that has been training canines to successfully locate scat from imperiled or listed carnivores specifically for conservation purposes.

Tortoises and Canines

Tortoises are land-dwelling, high-domed turtles that range over the Mojave and Sonoran deserts. The desert tortoise is an herbivore that may attain a length (upper shell) of 9–15 inches. This is a long-lived species that is slow to reproduce. The flattened front limbs are well adapted for digging burrows, and a single tortoise may dig a dozen or more burrows distributed over its home range. Tortoises are thought to spend up to 95% of their lives underground, enabling them to escape predators and desert temperatures that can exceed 140°F in summer or dip below freezing in winter.

The remains of a dead tortoise on the desert floor.
Tortoise skeleton
click to enlarge photo

When a tortoise dies, its shell may remain and leave behind clues to the age of the tortoise and, even in rare cases, the cause of death. By studying these shells, scientists attempt to estimate mortality rates. Based on such evidence coupled with other studies and monitoring activities, scientists have determined that populations of the Mojave Desert tortoise have been declining since the 1970s. Information on high-mortality rates resulted in the temporary federal emergency listing as "endangered" in 1989, and the official listing as “threatened” in 1990.

Natural predators of the desert tortoise include ravens, gila monsters, kit foxes, badgers, roadrunners, and coyotes, which prey on the 2–3-inch juveniles that have thin, delicate shells. Primary threats outside natural predators include illegal collection and vandalism by humans; destroyed habitat; upper respiratory tract disease; and loss of forage plants, or food, due to invasive plant species and competition with grazing livestock.

There is a long history of dogs working with humans to locate specific targets of interest, for example, drugs, explosives, accelerants, natural gas leaks, illegally poached wildlife, live humans, human remains, termites, and forensic evidence. Several recent high-profile searches where handler/dog teams were deployed include the Columbia space shuttle disaster and 911 terrorist attacks. In both instances, these missions involved the recovery of human remains. In the summer of 2003, dogs assisted in locating historical and previously unknown gravesites of the famed Donner Party in Truckee, California.

There are reasons why dogs are so successful at such tasks. Dogs possess a remarkable sense of smell and the ability to communicate what they smell to people. The sense of smell in the average dog is thousands of times more powerful than in humans. It follows that partnering dogs and humans is a natural approach to overcoming barriers that humans alone face in surveying the populations of the Mojave Desert tortoise. However, even with a keen sense of smell, not just any dog can become a “specialist.”

First, and most importantly, the dog must have a high level of “drive,” or motivation to work. Drive is often confused with high-energy, out-of-control behavior and considered a defect in a pet. Working dogs possess different types and combinations of drive—such as prey, play, hunt, and food. Cablk selects dogs for play and hunt drives only when seeking appropriate wildlife detection dogs. This combination of drives motivates the dog to continue searching for tortoises after long hours in the field and, upon finding one, immediately switch focus to the handler, who carries the toy reward. In this manner, the dog is fanatic about finding the tortoise, but after it does, it ignores the tortoise completely. Because tortoises are federally protected, ensuring their safety was a primary concern throughout the pilot project. Dogs that find tortoises must resist all temptation to interact with the tortoises directly, which can be difficult given the soccer-ball shape of adult tortoises!

To become a working dog, a canine must be fit, agile, and temperamentally sound enabling it to work in difficult terrain for long periods. For tortoise detection, both dog and handler must be athletes and possess a special ability to communicate clearly and consistently with each other.

Research Image Gallery

Click an image to zoom. Use your arrow keys to switch between images.

WDC scat dog Tsavo with his handler training in an open Montana field.

Tortoise dog Fin readies with one paw off the ground, poised to pounce on the ball the moment his handler tosses it in the air.

Fin lunges to chase the ball - earning his reward for a job well done.

Study Design

The overarching objective of Dr. Cablk’s study focused on assessing the efficacy and reliability of wildlife detection dogs to locate only live desert tortoises—not scat or carcasses—on the surface and in burrows under natural but controlled desert conditions. Reliability is defined as the number of times a dog encounters a tortoise and responds with its trained alert to indicate that a find was made. The trained alert is a behavior, such as sitting or lying down, that the dog performs to communicate to the handler when the target—in this case, a desert tortoise—is located. Reliability is not a measure of a dog’s ability to find tortoises; rather, it is a measure of the dog’s usefulness as a survey tool.

Data-gathering aspects of this study began with initial scent-recognition training for the dogs, conducted in Montana by WDC dog handlers. Dogs were trained on residual scent—scent that can be captured and remains on a surface even though the original source has been removed. Tortoise scent was captured by wiping sterile gauze pads only over the living, shedding parts of the front portion of the tortoise, including the soft skin of the front legs, armpits, neck, and face. Upper and lower shells or hindquarters were not used in preparing training aids to avoid incorporating the scent of scat. The gauze pads containing the residual tortoise scent—or target scent—are referred to as training aids. Blank training aids—gauze pads from the same box that contained no tortoise scent—were also prepared. Presenting the dogs with blank training aids as well as tortoise training aids helped teach the dogs it was tortoise scent and not gauze that the handlers wanted the dogs to find.

Following the initial phase of scent-recognition training, the dogs were transported to the DTCC in Las Vegas, Nevada. There they spent several days becoming acclimated to the dry desert air and myriad of different smells, as well as being introduced to live tortoises. The final data-gathering tasks centered on the research trials conducted at the DTCC to evaluate the effectiveness of canines to serve as survey tools and ultimately aid in the recovery of Mojave Desert tortoise populations.

Research Image Gallery

Click an image to zoom. Use your arrow keys to switch between images.

Wildlife dog Camas gives her handler the <i>shepherd-stare</i> to indicate that she has alerted on the target scent (marked with the pink flag).

This tortoise demonstrates a relaxed position, while it thermoregulates in the shade of a bush.

Research Trials

A series of experimental trials were conducted at the DTCC to evaluate the wildlife detection dogs in locating live desert tortoises and signaling finds to the handlers. These trials included separate surface and burrow searches, as well as clearance surveys (thorough canvases of defined areas to account for the total population of a given species), which were conducted to compare the effectiveness of both dogs and humans at finding desert tortoises.

A GIS (geographic information system) was used to generate random points in the study areas. Tortoises were tethered to stakes placed at these points through a ring epoxied on the hind area of their upper shells. Two-foot tethers prevented the tortoises from moving away from these locations but allowed the tortoise access to shade or sunlight. The tether design was approved by the U.S. Fish and Wildlife Service to make sure that stress to the animals was minimized.

Both dogs and handlers were fitted with GPS equipment to record their precise movements in one-second intervals. Special meteorological stations were designed to collect information about environmental conditions simultaneously with search activities. An intricate and extensive relational database was developed capable of tracking an immense volume of data, including changing locations of the dogs, handlers, and tortoises; dog/handler pairings; weather conditions; and physical descriptions of the tortoises.

According to Dr. Cablk, “This is an incredible 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 help lead to a reverse in the negative trend in tortoise populations.”

Research Image Gallery

Click an image to zoom. Use your arrow keys to switch between images.

Ring epoxied onto hind portion of upper shell with transmitter attached.

Fin is working off lead at the DTCC. Both dog and handler are fitted with GPS locators.

Dogs Fill Special Niche

As human development continues to encroach on the habitat of rare, threatened, and endangered species, rapid, accurate and cost-effective wildlife census techniques are becoming increasingly important. Desert tortoises are notoriously difficult to find, especially at low population densities and in thick vegetation or rough terrain. Scenting abilities of trained detection dogs have been well-documented in law enforcement, and these abilities also have been successfully applied to similar applications for other wildlife species. Drs. Cablk and Heaton’s study showed that trained wildlife detection dogs can effectively and reliably search out Mojave Desert tortoises—being significantly more effective than humans at finding tortoises camouflaged in shrubs, as well as hidden in burrows. The dogs in this study demonstrated an amazing overall detection accuracy of 90% —a result not achievable with humans alone . In addition, dogs located tortoises at distances of more than 60 meters (approximately 200 feet.)

Humans  typically do not find hatchlings like this one shown here. However, the  dogs in this study located this size entirely on the scent.
Hatchling found by Wildlife dogs
click to enlarge photo

Through comparative clearance surveys, this preliminary research also demonstrated that dogs are capable of locating more tortoises than humans thus resulting in better estimates of actual population sizes. The dogs were able to find tortoises in the smallest size classes whereas humans alone almost never locate newly hatched tortoises. Data on hatchling and juvenile tortoises will contribute significantly to the body of knowledge on the natural history of the desert tortoise and may shed light on the plight of tortoises as a threatened species. For this reason, dogs could be especially useful in filling a special survey niche in the future. These results suggest that wildlife detection dogs may greatly enhance tortoise population surveys, especially with regards to small tortoises that typically go undetected.

Dr. Mary Cablk, DRI Assistant Research Professor,
may be contacted through the following avenues.
Office 775.673.7371
2215 Raggio Parkway
Reno, NV 89512
This email address is being protected from spambots. You need JavaScript enabled to view it.


This website uses cookies
We use cookies to personalise content, provide social media features and  analyse our traffic. We also share information about your use of our site with our social media and analytics partners who may combine it with other information that you’ve provided to them or that they’ve collected from your use of their services. You consent to our cookies if you continue to use our website. DRI Privacy Policy >>