An artificial intelligence tool, developed by Queensland, researchers, has been able to identify the song of an endangered bird, helping scientists find the elusive feathered friend.
Eastern Bristlebirds have not been seen, nor their songs heard, in southeast Queensland in three years since the Black Summer bushfires which ripped through the state, as well as New South Wales and Victoria.
The Eastern Bristlebird is listed on the federal government’s website as “Endangered” in Queensland and NSW, and “Critically Endangered” in Victoria.
However, International Union for Conservation of Nature (IUCN) criteria suggests the northern population, with an estimated 30 individuals according to the Queensland government, is also critically endangered.
Conservation groups, government and researchers teamed up to try and find the rare bristlebird.
National bird conservation charity BirdLife Australia and environmental organisation Healthy Land and Water have been deploying acoustic recorders throughout the southeast Queensland bush. The sound captured by the microphones is analysed by Queensland University of Technology and QLD Department of Environment and Science researchers.
The recordings were screened by an AI model developed by Dr Lance De Vine from QUT’s School of Computer Science.
Five separate recorders detected a combined total of 350 Eastern Bristlebird calls in July and August of 2022.
Read more: Sharks and rays are at extinction risk, but climate change isn’t their biggest threat
“Passive acoustic monitoring is revolutionising how we survey soniferous biodiversity (animals that produce sound),” says associate professor from QUT’s Centre for the Environment Susan Fuller. “It is non-invasive, verifiable, and provides a permanent record of the environment.”
“We can record continuously which means we have more chance of detecting cryptic and low density or threatened species like the Eastern Bristlebird,” Fuller adds. “Now we have developed software tools to rapidly scan the big data generated from continuous passive recordings and hone in on a particular species.”
While sticking microphones in the wild to capture birdsong over a period of months is all well and good, researchers are overwhelmed with too many hours of recordings to listen to them all.
“We need solutions to go through the data quickly, searching for the species that we are interested in. This is where artificial intelligence comes in,” says QUT research assistant Callan Alexander.
“Basically, we transform the audio into an image, and then train a deep learning model to only search for images that we are interested in. In this case, we train our model to detect Eastern Bristlebird calls. The tricky part of this, is that Eastern Bristlebirds are notorious for having complex and highly variable vocal repertoires. So, we had to go through a lot of acoustic data manually to capture enough training data.”
Alexander adds that the Eastern Bristlebird has been brought to the brink of extinction by unsuitable fire regimes and invasive weeds. He is, however, hopeful that better understanding of distribution and numbers will help protect the birds.
“With more acoustic data to be analysed and further surveys to take place, we are very hopeful that the number of birds, extent of territories and evidence of breeding will be able to be determined,” he says.
Read more: The return of rare native songbirds after local extinction
Fuller is also optimistic that the technology can be used to aid in the conservation of other species. “The technology is improving rapidly and will certainly be used for more species in the future.”
Funding has also been accepted for the development of models for four other threatened species – Albert’s Lyrebird, Cascade Treefrog, Fleay’s Barred Frog, and Koala.
BirdLife Australia reports that the Black Summer fires are estimated to have burned over 97,000 square kilometres of native habitat. More than four billion animals are thought to have been killed, harmed or displaced, including many native birds.