A Bournemouth University PhD student is developing artificial intelligence to support the conservation of barn owls in Dorset.
Kavisha Jayathunge is working with Brian Cresswell, an electronics engineer with a background in biology, to create a system that can detect and count baby barn owls using sound alone.
The AI model can identify individual chicks based on subtle differences in their hissing calls. These differences are too small for humans to notice but can be recognised by the technology.
Unlike video cameras, which often disturb wildlife, this system allows for monitoring without entering nests. This could help researchers collect data from natural nest sites, not just from nest boxes.
Testing in a nest box in North Dorset has already shown that the AI can successfully identify three chicks based on their calls.
Strict rules apply, as barn owls and their nests are protected under Schedule One of the Wildlife and Countryside Act.
The long-term aim is to allow volunteers to use simple recording devices outside nests. Because barn owl hissing is loud, recorders do not need to be placed inside the nest to be effective.
The project could help scientists track chick behaviour and feeding, and eventually even measure hunger levels through sound.
Kavisha said the system could “make barn owl nest monitoring more scalable and less invasive” and hopes it will improve long-term outcomes for the species.