For 40+ years we've helped over 300,000 dog trainers just like you!

Learn more about Leerburg

$6.99 Flat Rate Shipping

Learn more
Ask Cindy Our Newsletter Free Catalog


Pilot study to investigate the use of conservation detection dogs to assist with geometric tortoise (Psammobates geometricus) surveys on Voelvlei Nature reserve.

Geometric tortoise surveys have been conducted for an extended period on Voelvlei Nature Reserve. Due to the fact that the tortoises hide away under shrubs, bushes or in burrows there is a perception that the current human conducted surveys are underestimating the population size. Evidence to support this was obtained during the recent fire in the area. 
In order to be able to manage the population of this endangered species better and understand the impact that potential threats are having on the geometric tortoises it is key that the surveying techniques are improved. The more accurate the surveying technique, the better the data obtained (i.e. population size, biological characteristics, habitat selection) to inform management decisions for this priority species. 
The concept of using dogs to assist with the geometric tortoise surveys has been proposed. The theoretical investigation into this has shown that this survey method can improve the quality and quantity of surveying data by increasing the detection rate of the biological target, decreasing the time and manpower required to conduct surveys and increasing the variation in size/age classes of the biological targets detected.
The project aim is to successfully implement this concept as outlined below.  
‘Conservation detection dog’ is an umbrella term for detection dogs trained to locate or discriminate biological targets in a natural setting for research or management application as opposed to dogs trained to find wildlife contraband in a law enforcement context. Dogs have been shown to be able to differentiate at least 10 different compounds without degradation in performance and become more efficient as new target compounds are learnt.
Surveys using conservation detection dogs have grown increasingly popular overseas in the UK, Europe, New Zealand, Australia and the USA and have started to gain popularity here in SA. Overseas they have been used to find a wide variety of items including invasive plant species, bat carcasses under wind turbines, scats of kit foxes, black bears, grizzly bears, bobcats, Pine Marten scats, dormouse nests and right whales in marine surveys.  Here they have been used to find cheetah scats, track down poachers, escaped tigers (Panjo), blue crane eggs etc. Working with dogs can increase the area surveyed as well as detection rate of survey targets. Rox Brummer from the EWT Limpopo Cheetah project has worked with Louise Wilson Director and General Manager for Wagtail UK LTD training two dogs to search for cheetah scat and has been to the Karoo to search for riverine rabbit in March 2011.
Spearheaded by Mary Cablk with Working Dogs for Conservation (NGO) the Desert Research Institute has put together a comprehensive report on the ‘Validation and Development of a Certification Program for Using K9s to survey Desert Tortoises’. The Mojave Desert tortoise (Gopherus agassizii) is listed as ‘Threatened” (US Endangered Species Act) and is in decline.  The use of dogs to survey this rare and cryptic reptile has been demonstrated to be safe, effective and cost efficient. Dogs were initially investigated for use as an alternative survey tool to humans who are limited to visual cues for finding tortoises that spend a majority of their time underground in burrows.  Dogs also increased the scope of the surveys by being able to find tortoises of varying size classes (very small constituting 30, 40, 50x2 and 80mm). Detection distances ranged from 0.5m to 62.8m.
As opposed to searching for scat, surveying for live animals presents additional challenges over other detection disciplines. Interfacing dogs with wildlife, particularly a listed species, generates a risk to the target that does not factor in surveying for inanimate objects. Tortoises may illicit a response when confronted by a dog which can increase risk. In contrast, should a dog ingest or destroy a scat only data is lost. With a live animal not only is data lost but harm may occur to the target species and a permit violation may ensue.
Criteria for a conservation detection dog:
A dog is driven (motivated) to improve his circumstances. Instinct!
Basic concepts of learning:

Trainers condition dogs to associate the odour of a target object with a highly-prized reward.
To work in this sort of environment you really require quite a particular type of dog and approach. You need:

The characteristics needed in a dog are an important first step in the labour-intensive process of assessing, training and eventually putting a dog to work. Many dogs can be trained to find scat. Fewer dogs have the right combination of drives and temperament to do live animals.
Training methodology and information.
There is a training protocol and programme outlined in Final Report. Validation and Development of a Certification Program for using K9s to Survey Desert Tortoises. ESTCP Project RC-200609. This method is used across the globe and entailsusing gauze swab pads scented from skin on front neck, head and legs area of tortoise as ‘bait’. Transition to captive live tortoises in field conditions where the tortoises were tethered and locations known then followed. Dogs did use scent & sight to locate tortoises. Minimum training period is 16 weeks full time.
Tortoises present a greater variability in scent than single molecule targets such as cocaine.
Size of tortoise located ranged from 89mm to 300mm.
“The handler ideally needs to be a biology graduate who understands sampling etc., rather than a junior level position. It needs someone with the confidence to stand up to rangers and researchers and say no if the dog is finished, or they are asking the impossible.” Rox Brummer.
Depending on size of area, more than one dog/handler team may be required in the long term.
Experimental results from data collected by dogs vary due to the quality of protocols used to train the dogs and the capabilities of the personnel involved in the research. It is particularly challenging to accurately design and execute a research study where dogs are trained for olfactory investigations and then used to collect data. Dogs learn independently often picking up on subtle cues from their handlers or observers who are unaware they are influencing the dogs’ responses. Great care is required to prevent cross contamination of target and non-target material. Dogs’ responses cannot always be validated by humans either quickly if at all. Researches are typically not trained nor do they necessarily understand that these behavioural and physical sensitivities exist. 
Three key principals in conservation detection dog success and assessment are:

Higher temperature will cause more evaporation and higher winds will transport the odour farther.
Dogs were able to learn tortoise scent through extraneous anthropogenic scent and located them from a long distance regardless of the tortoise’s level of human handling.
Dogs can detect as well as differentiate tortoise odour from all other non-tortoise scent. Dogs never alerted on tortoise carcasses, scat or urine during trials.

Exactly what constitutes tortoise scent that the dogs cue remains unidentified.  Variables that affect the search strategy include source strength, odour transport and dispersion by air movement.
The more target variants the dogs are trained on, the more unfamiliar variants they will correctly identify. It seems that rate of scent evaporation is effectively independent of tortoise size.

It has been shown that dogs have the ability to be trained on residual scent from a relatively small number of live tortoises kept in captivity in a highly unnatural environment, fed on a non-natural diet and maintained in unnatural enclosures with unnatural bedding material and then go on to find completely different and unrelated live tortoises maintained outdoors in their native habitat with minimal to no subsidy by humans.

Challenges in the field in South Africa.

Advice regarding the presence of more than one tortoise species in the field.

Louise Wilson:
“If you are wanting to just single out one of the species (endangered geometric tortoise) then I would train the dog on this tortoise and this tortoise species only.
It is not too much to ask for a dog to single out a specific species, as research suggests a dog can discriminate between single individuals also. A dog may generalise over many species of Tortoise, but ultimately the success of your project lies on the training/trainer that you use, the accurate and uncontaminated scents that you use, the dog breed that you use and the CT (continuation training methods) that you use and how regular.
Personally I would train the dog only on the Geometric Tortoise, this proves difficult due to limited availability of the Tortoise but are you able to get hold of one tortoise for a day? Every once and a while?”

Mary Cablk:

“I could envision scenarios, where the dogs could differentiate and also where they would generalize. This depends on training. If your goal is to have a species-specific canine, then training only on the one species first, followed by what we call "proofing" off the sympatric species would be the progression. If it is not possible to get training aids or actual animals to train the dogs on then the best, perhaps only, option is to hope that you can reinforce the generalization in the field. So what that means is, train the dogs on whatever tortoises you have and when they get to the field setting, reward them on all tortoises.

During the time that the dogs are generalizing in the field, which is an unknown amount of time, you'll have to remember that the data you get from the dogs will not necessarily be accurate on the new species. You'll need an experienced handler to work the dogs in this first attempt to recognize the opportunity to reinforce the new tortoise odor in the field. This is because we just don't know if the dogs will immediately recognize generic 'tortoise' or if they will identify some overlap in odor signatures between what they've been trained to and this new animal odor.

If you train the dogs on multiple other tortoise species then the likelihood that they will generalize in the field to a third or fourth tortoise species should be higher, assuming you put one trained alert for all target tortoise species.

I'd need more information on what happens in the field before making a determination on whether or not it is valuable to extinguish the alert on the non-endangered species after they have generalized the geometric tortoise.”

Project objectives.


The team.
Individuals that will be actively working to get this off the ground: Vicki Hudson, Natalie Haywood, Bernard Wooding: conservation manager at Elandsberg Nature Reserve.
Advisory group (local): Guy Palmer, Ernst Baard, Arnelle van Noie, Lee Saul, Retha Hofmeyer.
Advisory group (national & international): Louise Wilson, Rox Brummer, Mary Cablk, Dave Harris.

Materials and methodology.
The training methodology outlined in the Final Report. Validation and development of a Certification Program for Using K9s to Survey for Desert Tortoises protocol will be followed. It is downloadable from Vicki Hudson will test this methodology on her retired working dog, Bernard Wooding on his dog and Natalie Haywood on her dog. Bernard will need to apply for the necessary permit from CapeNature Head Office to carry out this work. Louise Wilson from Conservation Dogs: Wagtail UK ( has indicated that she is willing to come out and assist us to set up this programme. Transport and accommodation are able to be provided in lieu of her services at this stage. Once this has been setup successfully the long term view would be to extend the training to increase the number of certified dog/handler teams that operate within the project on a voluntary basis. Standard Operating Procedures for Conservation Dog/Handler teams have already been drawn up and can be adapted and utilised.

The training will take place at Elandsberg initially as they currently have the ideal circumstances for this in the form of approximately 50 captive geometric tortoises in 2 large camps (-/+ 2 Ha). Once the dogs are suitably imprinted to the scent and alerting reliably the surveys in the field can begin i.e. on Voelvlei Nature Reserve. 

Legal requirements.
CapeNature personnel participating in the project do not require permits. However private individuals i.e. Bernard Wooding would require a permit. (Pers comm Deon Hignett)

Training a dog to detect scat is much easier and it’s even easier to detect a carcass compared to a live animal. But the instinct of the dog, when trained properly, the dog is able to adapt to indicate on living animals. Concern about conservation detection dogs would be if the wildlife in question was to bolt and run away at any point when the dog is searching, this would cause a prey drive reaction – i.e. the dog would chase it due to instinct- however a tortoise is unlikely to trigger this!

The use of conservation detection dogs should be a non-invasive method of surveying and retrieving or interacting with the target is out. The dog should be looking to the handler for reward not the target - which he should be impartial to and not interact with in any way. Irresponsible actions can ruin this type of work completely & unjustly give the organisation, dogs & process a bad name. Public perception is an important factor – even more so when working for a government organisation.  

Tortoise detection dogs have been used before and proved to be highly successful. If done it must be carried out in a professional, scientific and legally legitimate manner, existing expertise & proven methods must be implemented with long term commitment  It will require time, expertise and cost to set it up.

The project should be spearheaded by CapeNature with partnerships from various sectors including private landowners from contract stewardship sites in the area with geometric tortoise populations.

There is scope in the conservation field in the province to expand this concept to maximise the working ability of those involved. With the current explosion of wind farm developments this methodology should be looked at when addressing the methodology for M&E of impact of wind farms on bat populations. Location of leopard scat could be another type of detection work that could benefit from this type of project.

Way forward.


2012 – Set up program, trial methodology, imprint odour, conduct trial surveys in situ.
2013 – Conduct surveys in field.
2014 – Conduct surveys in field
2015 – Assess & evaluate pilot study. Document and publish findings.

Bio3. A biodiversity, research and information systems consultancy. Portugal.

Cablk, M.E.  and J.S. Heaton. 2006.  Accuracy and reliability of dogs in surveying for Desert
Tortoise (Gopherus agassizii). Ecological Applications. 16(5):1926-1935.

Cablk, M.E., J.C. Sagebiel, J.S. Heaton and C. Valentin. 2008. Detection distance: A quantitative
analysis of how far away dogs detect tortoise scent and follow it to source.   Sensors. 8(4):2208 -

Cablk, M.E., Harmon, R. 2011. Final Report. Validation and development of a Certification Program for Using K9s to Survey for Desert Tortoises. ESTCP Project RC-200609.

Department of Conservation: New Zealand. October 2011. Conservation Dog/Handler team. Standard Operating Procedure.  Retrieved from

Heaton, J.S., Cablik M.E., Nussar K.E., Esque T.C., Medica P.A., Sagebiel J. C., Francis S.S. 2008. Comparison of effects of Humans versus Wildlife-Detector Dogs. The Southwestern Naturalist 53 (4):

Nussear, K.E., T.C. Esque, J.S. Heaton, M.E. Cablk, K.K. Drake, C. Valentin, J.L. Yee, and P.A.
Medica. 2008. Are wildlife detector dogs or people better at finding tortoises (Gopherus
agassizii)? Herpetological Conservation and Biology. 3(1):103-115. 

Paula, J., Leal, M., Silva, M.J., Mascarenhas, R., Costa, H., Mascarenhas, M., (2010). Improvement of bird and bat carcasses detection on wind farms using wildlife research dogs. Wind Wildlife Research Meeting VIII, Denver, Colorado, USA, 19-21 October 2010.

Paula, J., Leal, M. C., Silva, M. J., Mascarenhas, R., Costa, H. & Mascarenhas, M. 2011. Dogs as a tool to improve bird-strike mortality estimates at wind farms. Journal for Nature Conservation: 19(4): 202-208.

Reed, S.A, Bidlack A.L, Hurt A., Getz, W. 2010. Detection Distance and Environmental Factors in Conservation Detection Dog Surveys. Journal of Wildlife Management 75(1):243-251.

Retrieved from Sniffer dogs being used to track down cheetah. July 2010.


Rox Brummer. EWT Limpopo Carnivore Project. South Africa . Email contact.
Louise Wilson. Wagtail UK; Conservation Dogs. England. Email contact.
Mary Cablk. Associate Research Professor. Desert Research Institute. Reno USA. Email contact
Bernard Wooding: Elandsberg Nature Reserve. Telephone contact.
CapeNature: Natalie Haywood, Lee Saul, Arnelle van Noie. Email and telephone contact.

20% off select DVDs, streaming videos, and online self-study courses through Sunday, April 11, 2021 at 11:59 PM CT