Expert Tracking Revolution: AI Poised to Transform Wildlife Research and Conservation

A pioneering artificial intelligence (AI) system is poised to revolutionize the field of wildlife research, offering hope to researchers grappling with the challenges of tracking elusive creatures like Namibia's brown hyenas. This species, more closely related to mongooses than to dogs, has proved particularly difficult to study due to their nocturnal habits and wariness of camera traps.

Marie Lemerle, a dedicated researcher with the Brown Hyena Research Project, knows these challenges all too well. With experiences ranging from hyena cubs destroying camera equipment to the difficulty of monitoring these solitary mammals, Lemerle's work could be significantly bolstered by a new tracking tool that's currently being developed.

The US-based non-profit organization WildTrack, led by conservationist Zoe Jewell, teamed up with Lemerle to refine an AI system that can identify animals based on images of their footprints. The technology, which has been under development by WildTrack for the past 13 years, already has the capability to recognize 17 different species, including leopards, lions, and rhinos.

For the past five months, Lemerle has been gathering vital hyena footprint data at Baker’s Bay, a hub for Cape fur seals and a hunting ground for brown hyenas. By photographing hyena tracks with a standard 30-centimeter ruler for scale, Lemerle is aiding in the creation of a robust reference library that will train the AI in footprint identification.

At Duke University, the WildTrack team analyzes the footprints in great detail, examining up to 120 distinct measurements to distinguish between individual animals. They look for subtle differences such as the angles between the toes, and even the imprints of injuries, which many brown hyenas sustain during fierce competition at feeding grounds like Baker’s Bay.

The AI tool promises to be a transformative addition to traditional study methods. It could offer researchers like Lemerle insights into the movements and behaviors of individual hyenas without the need for direct observation. This has enormous potential to enhance our understanding of these mysterious creatures and better inform conservation strategies.

WildTrack's ambitions for their AI system extend beyond the realm of academic research. With an estimated population of fewer than 3,000 adult brown hyenas in Namibia and less than 10,000 in all of Southern Africa, the species is considered near threatened. Factors such as vehicle collisions and killings by livestock farmers due to perceived threats have exacerbated their plight. The WildTrack AI tool and the associated smartphone app could serve as evidence to exonerate brown hyenas from suspected livestock attacks, potentially decreasing the number of retaliatory killings.

Wesley Gush, a graduate student at the University of Pretoria, has underlined the significance of such a tool, emphasizing the hidden nature of brown hyenas and the resultant underestimation of their population numbers. The Automated Tracking System promises not only to aid researchers like Gush in their studies but also to play a crucial role in the preservation and protection of one of Africa's most cryptic large carnivores.

The hope, as expressed by Jewell and conservation enthusiasts, is that the AI model created for studies like Lemerle's can be applied across the globe to safeguard brown hyenas and a plethora of other vulnerable species.