Researchers have created a non-invasive technique using footprints to identify nearly identical small mammal species, achieving up to 96% accuracy in tests on sengis. This method promises to enhance monitoring of these vital environmental indicators without relying on costly DNA analysis. The approach was developed to detect early signs of ecosystem damage through subtle differences in animal tracks.
Small mammals, often overlooked compared to charismatic species like lions or pandas, serve as crucial sentinels for environmental health. Their populations shift rapidly in response to habitat changes, making them ideal for early detection of biodiversity threats. However, many of these animals belong to cryptic species that are visually indistinguishable, complicating conservation efforts.
A team led by Dr. Zoë Jewell from Duke University's Nicholas School of the Environment has addressed this issue with a innovative footprint analysis system. Published in Frontiers in Ecology and Evolution, the study focused on two sengi species: the Eastern Rock sengi and the Bushveld sengi. Despite their similar appearances, these elephant-shrew relatives inhabit distinct environments—one favoring rocky areas and the other sandy terrains—and face unique pressures from climate and land use changes.
"It's often only possible to distinguish between cryptic species using DNA, which can be slow, invasive, and costly," Jewell explained. The new method trains computer models on high-resolution images of footprints, capturing subtle variations in foot shape and size. Field tests occurred in South Africa's Telperion Nature Reserve and Tswalu Kalahari Reserve, where 19 Eastern Rock sengis and 18 Bushveld sengis were gently captured using baited traps stocked with oats, peanut butter, and Marmite.
The animals walked across charcoal-dusted paper in collection boxes, leaving tracks that were photographed and measured via morphometry software. Nine key features from front footprints proved most discriminatory, yielding 94% to 96% identification accuracy on unseen data. Notably, some Eastern Rock sengis appeared in Tswalu, beyond their typical range, highlighting the need for such tools to track shifting distributions.
Jewell emphasized the broader potential: "Small mammals exist in almost every ecosystem on the planet, and our tech is flexible enough to adapt to every one." By offering an ethical, affordable alternative to invasive techniques, this approach could routinely assess ecosystem integrity and prevent silent declines in biodiversity.
