Post provided by Annegret Grimm-Seyfarth

Border Collie Zammy is trained to search for Eurasian otter scat and pond, alpine and great crested newts. Photo: André Künzelmann.

For those not directly working with them, using wildlife detection dogs always sound like a new fancy idea that should be tested somehow. However, this method is neither new nor rare, and people working with wildlife detection dogs often call them their best method in finding their target species. In this post, Annegret Grimm-Seyfarth discusses her paper ‘Detection dogs in nature conservation: A database on their worldwide deployment with a review on breeds used and their performance compared to other methods’, which shows the broad and worldwide applications of wildlife detection dogs.

From otter scat monitoring to detection dogs

About eleven years ago, I took a practical Bachelor’s course where we had been trained to find and collect Eurasian otter scats. This was part of a bigger research project and scats were analysed genetically afterwards. We searched for a week and obviously only found exposed scats. However, the truly disappointing part was that over a third of all the scats the researchers and we found belonged to syntopic species, most of them to the American mink. These species had also fed on fish and thus their scats were indistinguishable for us, visually and olfactorily. I thought there must be a better way of doing this, and as a dog-person, I immediately thought of a detection dog.

I started reading around and discovered that this was indeed a possible monitoring method, called scat detection dog. I spent the next two years reading and collecting papers before I finally got my own young and overly active dog from an animal shelter (I had dogs before, but obviously they were selected based on different criteria). While training this dog together with some professional trainers, I couldn’t stop reading about how others trained them; which dogs they used, what these dogs were trained on and where they have been deployed. And I was impressed by what these dogs are able to do, to learn, to work on.

The surprising increase in literature

Border Collie Zammy is sent to search for newts in a shrub area by Annegret Grimm-Seyfarth. Photo: Daniel Peters.

So far, literature seemed not to be overly abundant, so I started collecting everything in a database. In 2016, I had trained four dogs as scat detection dogs and two of them became inevitable for otter scat detection in the research projects I work in. I gave a symposium presentation proudly presenting the results of our work and 315 publications about different kinds of wildlife detection dogs. This was quite a number for a not-well-known method, especially in Europe. How many more papers can there be? In 2018, another presentation with “first results” included 627 publications, and at an international conference in 2019, it was up to 751 references. Meanwhile, I had finished my PhD and successfully established the use of wildlife detection dogs for various species in our lab. But more importantly, I had convinced two other people who were as crazy about wildlife detection dogs as myself to help me with what was starting to become a crazy adventure about literature searches, translations, database entries and analyses.

At some point, when we figured that new publications seemed to appear weekly and we had covered the past quite well (thanks to our library!), the three of us decided that we have to sum it up in a review article. We ended up with 1220 publications from 1930 to October 2020, describing 2464 individual cases of specific breeds searching for specific species in a specific country. We focused on scientific publications, but as wildlife detection dogs were frequently used for conservation or management purposes without a scientific research project behind, we also included publications from and about them in books, popular science or newspaper articles. Indeed, the increase in publications was real: Since 2000, the annual number of scientific publications increase by 161% per year. In other words, back in 2010, we could find 31 scientific publications published that year. In 2019, we found 49. When considering all publications, we could find 40 in 2010 and 97 (!) in 2019. We also found a significant annual increase for book, popular science and newspaper publications although we only included those studies not mentioned in a scientific publication and without double-counting publications about the same study. This clearly shows that wildlife detection dogs are not only increasingly used in research, but also in conservation, wildlife management and landscape planning. If Richard Henry from Resolution Island, New Zealand, the pioneer who used detection dogs for conservation purposes, had known how ground-breaking his ideas were back in the 1890s, he could have asked for more honour and money during his lifetime, instead of the loneliness and depression he had struggled with.

Worldwide deployment of wildlife detection dogs

Australian Shepherd Fine is searching for hedgehogs to help co-author Anne Berger. Photo: Katja Krauß.

Even if the pioneer’s idea was not immediately taken up, wildlife detection dogs have a long and enduring history in New Zealand. They were divided into “protected species dogs” trained to detect rare species, and “predator dogs” trained to detect invasive alien predators for eradication programs. With the worldwide increase in the use of wildlife detection dogs, their tasks have changed and become much more diverse. Nowadays, they also include the detection of pests (e.g., invasive plants, arthropods, fungi), traces (e.g., scat, hair), carcasses (e.g., in wind parks, under power lines, poison monitoring) and animal quarters (e.g., dens, roosts, nests). The use of detection dogs for conservation rather than hunting purposes has been known from the USA since the 1920s and from the UK, Scandinavia, Russia and South America since the 1960s. These early studies already showed a variety of target species, from game birds and mammals to rodents, big cats, seals, snakes and turtles. Pioneer studies for scat detection dogs stem from the late 1970s and early 1980s, but only after advances in genetics and endocrine analyses in the 1990s, their use increased exponentially (see here for a historic overview).

By now, we could track reports for 62 countries and 408 animal, 42 plant, 26 fungi and 6 bacteria species, numbers that are going to increase in the next years. Altogether, 108 FCI-classified and 20 non-FCI-classified breeds have worked as wildlife detection dogs. In line with the history of different dog breeds and the conservation and wildlife management history in different countries, we found that certain breeds have been preferred on different continents and for specific tasks and targets. However, they were not generally better suited for detection tasks than others, which contradicts previous assumptions.

Advantages and limitations of wildlife detection dogs

Generally, wildlife detection dogs worked more effectively than other monitoring or survey methods. For each species group, regardless of the dog breed, detection dogs were better than other methods in 88.71% of all cases and only worse in 0.98%. We could detect a few specific target-breed-combinations, which less often outperformed other monitoring methods. For example, we found evidence that terriers were less suited for mammal detection. However, they were often used in eradication programs where a range of different methods is necessary for success, and training the predator dogs was sometimes not species-specific enough. Apart from eradications, it was only for arthropods that Pinshers and Schnauzers performed worse than other breeds. However, the sample size for this FCI group was much smaller than for most others, and single studies could thus drive such results much stronger. Finally, we found that wildlife detection dogs, regardless of the breed, were much less likely to outperform other methods for mono- and dicotyledons. While the use of plant detection dogs has some undeniable advantages when detecting small or underground plants, detection dogs run into problems when entering a so-called “scent pool” with many individuals of the target plant species and can then no longer distinguish among individual plants.

Our current wildlife detection dogs and their main target species: Bagheera, Australian Cattle Dog, otter scat detection (left); Fine, Australian Shepherd, hedgehog detection (middle); Zammy, Border Collie, newt detection (right). Photos: Annegret Grimm-Seyfarth and Anne Berger.

Summary

In summary, we found that any breed can be trained as a wildlife detection dog. However, selecting the most suitable dog for the task and target may speed up training and increase the chances of success. Apart from dog selection, we found that excellent training, knowledge about the target species’ density and suitability for detection dog work, and an appropriate study design were the major obstacles when detection dogs were not superior to other monitoring or survey methods. Moreover, we confirmed findings from case studies that a suitable area, habitat and weather are crucial for detection dog work. When these factors are taken into consideration, wildlife detection dogs can be an outstanding monitoring method. For the three of us, all our detection dogs became an essential part of our fieldwork, almost guaranteeing that we obtain comprehensive data essential for all further analyses.

To read the full Early View Methods in Ecology and Evolution article ‘Detection dogs in nature conservation: A database on their world‐wide deployment with a review on breeds used and their performance compared to other methods’, click here.