When a team of scientists listened to audio recorded underwater off islands in central Indonesia, they heard a sound that sounded like a bonfire crackling.
It was a lively coral reef, according to a study published last month by scientists at British and Indonesian universities. They used hundreds of such audios to train a computer program to monitor the health of a coral reef by listening to it.
A healthy reef makes a complex “sizzling sound like a campfire” because of all the creatures that live on it, said lead researcher Ben Williams. A degraded reef is quieter.
The artificial intelligence (AI) system analyzes data such as the frequency and volume of the audio sound and is able to determine with at least 92% accuracy whether the reef is healthy or degraded. The information is contained in the team’s study, published in the journal Ecological Indicators.
Scientists hope this new artificial intelligence system will help conservation groups around the world more efficiently monitor the health of reefs.
Coral reefs are under pressure from man-made carbon emissions, which have raised ocean surface temperatures by 0.13 degrees every decade and have increased their acidity by 30% since the industrial age.
According to the Global Coral Reef Monitoring Network, about 14% of the world’s reefs were lost between 2009 and 2018 — an area 2.5 times the size of Grand Canyon National Park in the United States.
Despite covering less than 1% of the ocean floor, coral reefs feed more than 25% of marine biodiversity, including turtles, fish and lobsters, which is why they are heavily targeted by the global fishing industry.
Indonesian conservationist Syafyudin Yusuf, a professor in the faculty of marine sciences at Hasanuddin University, said the research will help monitor the health of reefs in Indonesia.
Scientists also hope to collect underwater recordings of reefs in Australia, Mexico and the Virgin Islands to help them assess the progress of reef restoration projects.
