Conservation news

For monitoring mammals, eDNA boasts ‘killer feature’ over other methods

Grizzly bear

Banner image: A grizzly bear (Ursus arctos) recorded on camera trap in British Columbia by researchers comparing the efficacy of eDNA and camera traps to detect terrestrial mammals. Image ©Robin Naidoo/WWF

  • Mounting evidence suggests that the fast-developing tool of eDNA could be a game-changer for terrestrial mammal monitoring.
  • A new study demonstrates that eDNA analysis of stream water can reveal the diversity of terrestrial mammals in a large landscape as effectively as camera trapping and for a fraction of the cost.
  • Traditional mammal survey methods can be time-consuming, expensive and far from failsafe; eDNA is a reliable and comparatively inexpensive way for conservationists to gain a snapshot of an ecosystem’s mammal fauna, scientists say.
  • It could also have a big impact on conservation, since eDNA data allows timely decisions on which species to prioritize and which areas to protect.

A European meadow viper (Vipera ursinii), one of the world’s most threatened snakes, glides undetected between blades of grass in an Alpine meadow. It slips into a rocky outcrop and disappears. On its trail is scientist Arnaud Lyet, who is trying to estimate the viper’s population size; but this particular snake slithers from view unrecorded. Lyet ponders a question that routinely confronts many conservationists studying scarce and elusive species: does not seeing something mean it isn’t there?

“This question of detectability is really important,” Lyet tells Mongabay. “One of the challenges that wildlife ecologists and conservationists face is that the species we are interested in are difficult to observe, so information is difficult to collect in order to understand what actions you need to take to protect them.”

Conservation efforts are essentially based around knowing what lives where, so finding ways to improve wildlife monitoring accuracy is vital.

Over the past 10 years, Lyet has shifted his focus from snakes to mammals. He is now examining wildlife populations for WWF using environmental DNA (eDNA), a fast-developing tool that is revolutionizing the monitoring of aquatic life.

Mounting evidence suggests the technique could be a game-changer for terrestrial mammal monitoring too. Various studies showcase it as a reliable and comparatively inexpensive way for conservationists to gain a snapshot of an ecosystem’s mammal fauna.

The use of eDNA to detect species depends on the fact that as animals move through their environment, they shed traces of DNA: skin, hair, feces, saliva. Since a lot of this material is washed from land into streams through rainwater or enters watercourses directly when terrestrial animals bathe, hunt or drink, scientists can sample eDNA from waterways to assess wildlife presence over the wider catchment. A syringe is used to push water through a filter, which traps organic material containing the DNA. This residue is carefully stored to avoid contamination and sent off for laboratory analysis.

In the lab, scientists extract the DNA sequences and apply a technique called metabarcoding, which attaches a primer to amplify the DNA of a target species group, such as mammals, birds or fish. The distinctive DNA sequences are then compared against known species profiles in a reference database.

Arnaud Lyet deploys eDNA equipment to sample the lower catchment of the South Chilcotin mountains, British Columbia, Canada. Image ©WWF

Conclusive results

In his search for improved monitoring methods, Lyet recently led a team that demonstrated that eDNA analysis of stream water can reveal the diversity of terrestrial mammals in a large landscape as effectively as camera trapping. The findings were published in Scientific Reports in May.

At the outset of the fieldwork in 2018, Lyet said he wondered whether they would find anything at all in the water, especially in some of the smaller streams. But in the end, he was convinced of the effectiveness of eDNA. “I was so relieved when I saw the first results showing the amount of species that were detected … I did not expect to get a result that was so conclusive.”

Traditionally, scientists have monitored mammal species by catching and tagging animals or placing camera traps throughout an animal’s habitat. But these techniques are often time-consuming, expensive and far from failsafe: it is up to chance whether or not a tiger prowls in front of a camera in the depths of a forest.

Sampling the eDNA in river water is a comparatively simple and low-cost method to gain an overview of which species, including rarities, are present over broad spatial scales, according to Lyet. It could also have a big impact on conservation, since eDNA data allow timely decisions on which species to prioritize and which areas to protect.

In 2018 and 2019, the researchers sampled large volumes of water from streams across a catchment area encompassing more than 1,200 square kilometers (460 square miles) in the South Chilcotin Mountains, British Columbia, Canada. They also deployed 57 camera traps across the same landscape.

They found that eDNA methods detected 25% more terrestrial mammal species than camera trapping: 35 and 29 mammal taxa, respectively. Water sampling closer to the base of the catchment yielded most detections. Traces of mammals, including red squirrels (Tamiasciurus hudsonicus), mule deer (Odocoileus hemionus), grizzly bears (Ursus arctos) and wolverines (Gulo gulo), were detected in water samples, including several bat species that went undetected by cameras. Moreover, the team calculated that eDNA methods were a fraction of the cost.

Robin Naidoo, a co-author of the new study, attaches a camera trap to a tree in British Columbia, Canada. Image ©Robin Naidoo/WWF

Mounting evidence

Similar studies confirm the proficiency of eDNA in detecting mammals compared with traditional monitoring methods. Besides stream water, studies targeting mammals have retrieved eDNA from samples of soil, sediment and even air.

One study by scientists at Stanford University in 2020 analyzed eDNA from soil samples collected along wildlife trails through woodlands, riverbanks and grasslands. The eDNA identified all mammals that nearby camera traps had regularly recorded during the previous four years. In addition, eDNA indicated the presence of a number of small mammals, including bats and voles, that had rarely, if ever, been recorded on camera.

“For me, the killer feature of eDNA is that you can detect species from a very broad range of taxa,” Kevin Leempoel, lead author of the Stanford study, who is now at the Royal Botanic Gardens, Kew, U.K., told Mongabay in an email. Using eDNA in the soil samples, he identified reptiles, arthropods, worms and birds.

According to Lyet, other projects have extracted eDNA from pawprints in snow. By analyzing snow tracks, a WWF-led pilot study revealed not just polar bears, but also retrieved the DNA of a seal that a bear had hunted and that of a gull that had feasted on the carrion.

In the Peruvian Amazon, a further WWF-led study found eDNA results to be comparable to other mammal survey methods, including camera trapping, mist netting for bats, and small mammal trapping.

Another study in the U.K. used eDNA to detect a suite of mammals, including badgers (Meles meles), pine martens (Martes martes) and water voles (Arvicola amphibius), the last of which camera traps failed to pick up.

Nevertheless, eDNA analysis does have limitations when compared to camera traps. It can only tell if a species is present in a sample; it cannot give detailed information on species abundance or whether animals are adult or juvenile, male or female. Neither can eDNA give accurate information about the spatial distribution of a species, as it is difficult to determine the source of the DNA in a sample; this is especially true for water samples.

“The range of studies that can be conducted with cameras is broader,” Leempoel said. “[But] eDNA is already a great complement to camera trapping because it greatly helps the detection of small mammals … Not only do small mammals rarely trigger cameras, unless we specifically look for them, but identifying them to species level on these pictures is very difficult.”

Wolves (Canis lupus) were detected well on camera traps and eDNA detected Canis species in a total of 32 samples but could not distinguish between wolves, coyotes and domestic dogs. Image ©Robin Naidoo/WWF

Build global capacity

In the tropics, eDNA has been successfully used to catalog mammals in a range of locations, including the Peruvian Amazon, the Atlantic Forest in Brazil, Colombia, and French Guiana. Such studies have found a diversity of species, from giant anteaters (Myrmecophaga tridactyla), armadillos and tapirs, to jaguars (Panthera onca), bats and night monkeys (Aotus spp.).

But detecting incredibly rare rainforest mammals sometimes requires extra ingenuity. In Vietnam’s Annamite Mountains, scientists recently used DNA extracted from leeches to study one of the region’s most elusive mammals, the Annamite striped rabbit (Nesolagus timminsi), whose population is believed to have halved since 2008. By sucking the blood of a variety of rainforest mammals, leeches essentially sample DNA throughout their habitat. Collecting leeches and sequencing the rabbit DNA in their bodies enabled the researchers to profile the Annamite striped rabbit’s genetic diversity.

Thanh Nguyen, a doctoral candidate at the Leibniz Institute for Zoo and Wildlife Research in Germany, who led the research, said that detection of rare species during wildlife monitoring is often imperfect. “A combination of multiple methods, such as camera trapping and different eDNA sources can be very helpful,” he told Mongabay in an email. “eDNA is still new, and only wider application will inform us on its effectiveness, but it certainly provides scientists and conservationists a new tool to detect and monitor these rare species.”

At present, the biggest obstacle to widespread adoption of eDNA analysis in the tropics is the development of a comprehensive reference database against which to identify species, according to Allan McDevitt of the University of Salford, U.K., who led the aforementioned studies in the U.K. and Brazil.

“In general, because tropical regions tend to be more speciose and sometimes understudied in comparison to more temperate regions, we can lack appropriate reference sequences in databases to accurately [assign] DNA sequences to a particular species,” he told Mongabay in an email, adding that gradually, more reference material is being gathered.

McDevitt said other monitoring methods are still vital. “For mammals, wide-ranging and generally solitary species such as carnivores are not detected as effectively using eDNA, so species ecology is important. A combination of techniques, such as camera trapping and eDNA, is an optimal approach to capture the majority of mammalian species in an area.”

Abigail Hehmeyer, a co-author of the new study, filters stream water to trap organic material containing DNA. The sample will be sent to a laboratory for analysis. Image ©WWF

Another challenge is that DNA extraction and sequencing is only possible in state-of-the-art laboratories. Many countries in the tropics that would benefit enormously from eDNA’s rapid and cost-effective biodiversity assessments lack such facilities. The current need to ship samples to laboratories overseas inevitably heightens costs. The situation is, however, improving: “There are now labs in Brazil that can do these types of analyses and other countries are moving in the same direction,” McDevitt said.

“If we want [eDNA] to become mainstream and used on a large scale, we need to build capacity around the world, so that people can collect samples in the right way, analyze the samples in their countries, and use the results for their own conservation and monitoring purposes,” Lyet said.

Programs to boost the global capacity for eDNA monitoring are already in the pipeline. For example, a partnership between WWF, ETH Zurich and research laboratory SPYGEN plans to dispatch mobile laboratories, conveniently sized as shipping containers and equipped to process eDNA samples, to countries in need. The program will begin in Colombia later this year to study the impacts of forest degradation and recovery on terrestrial biodiversity.

While eDNA will not completely replace traditional methods of monitoring terrestrial mammals, it can be used concurrently to gather high-quality data about what species are around. Furthermore, sampling can be conducted without specialized skills, thereby empowering local communities and organizations to take control of biodiversity monitoring.

Lyet says he is confident that that eDNA technology will improve and eventually become accessible to all, so that species and ecosystems can be monitored on a global scale: from bacteria and fungi to plants and animals.

Ultimately, he says, this will allow for more effective biodiversity stewardship. “Being able to monitor biodiversity as a whole through eDNA can give us better understanding to help us to make better decisions for conservation.”

Banner image: A grizzly bear (Ursus arctos) recorded on camera trap in British Columbia by researchers comparing the efficacy of eDNA and camera traps to detect terrestrial mammals. Image ©Robin Naidoo/WWF


Lyet, A., Pellissier, L., Valentini, A., Dejean, T., Hehmeyer, A., & Naidoo, R. (2021). eDNA sampled from stream networks correlates with camera trap detection rates of terrestrial mammals. Scientific Reports, 11, 11362. doi:10.1038/s41598-021-90598-5

Clare, E. L., Economou, C. K., Faulkes, C. G., Gilbert, J. D., Bennett, F., Drinkwater, R., & Littlefair, J. E. (2021). eDNAir: Proof of concept that animal DNA can be collected from air sampling. PeerJ, 9, e11030. doi:10.7717/peerj.11030

Leempoel, K., Hebert, T., & Hadly, E. A. (2019). A comparison of eDNA to camera trapping for assessment of terrestrial mammal diversity. Proceedings of the Royal Society B: Biological Sciences287(1918), 20192353. doi:10.1101/634022

Mena, J. L., Yagui, H., Tejeda, V., Bonifaz, E., Bellemain, E., Valentini, A., … Lyet, A. (2021). Environmental DNA metabarcoding as a useful tool for evaluating terrestrial mammal diversity in tropical forests. Ecological Applications, 31(5). doi:10.1002/eap.2335

Sales, N. G., McKenzie, M. B., Drake, J., Harper, L. R., Browett, S. S., Coscia, I., … McDevitt, A. D. (2020). Fishing for mammals: Landscape‐level monitoring of terrestrial and semi‐aquatic communities using Edna from riverine systems. Journal of Applied Ecology, 57(4), 707-716. doi:10.1111/1365-2664.13592

Sales, N. G., Kaizer, M. D., Coscia, I., Perkins, J. C., Highlands, A., Boubli, J. P., … Mcdevitt, A. D. (2020). Assessing the potential of environmental DNA metabarcoding for monitoring neotropical mammals: A case study in the Amazon and Atlantic Forest, Brazil. Mammal Review, 50(3), 221-225. doi:10.1111/mam.12183

Coutant, O., Richard-Hansen, C. R., De Thoisy, B., Decotte, J., Valentini, A., Dejean, T., … Brosse, S. (2020). Amazonian mammal monitoring using aquatic environmental DNA. Molecular Ecology Resources21(6), 1875-1888. doi:10.22541/au.160356435.56628486/v1

Nguyen, T. V., Tilker, A., Nguyen, A., Hörig, L., Axtner, J., Schmidt, A., … Fickel, J. (2021). Using terrestrial leeches to assess the genetic diversity of an elusive species: The Annamite striped rabbit Nesolagus timminsi. Environmental DNA, 3(4), 780-791. doi:10.1002/edn3.182

Feedback: Use this form to send a message to the author of this post. If you want to post a public comment, you can do that at the bottom of the page.