Archiving the orchestra of the ocean

We tend think of the ocean as quiet place. We picture the enveloping calm that enfolds us we duck beneath the surface as an inherent property of the sea – the hubbub of the terrestrial world slips away and is replaced by a profound tranquility.

But in fact, nothing could be further from the truth. The ocean is humming – and crackling, popping, growling, grunting – with the sounds of its inhabitants.

All marine mammals are believed to emit noises, which range from the mournful melodies of humpback whales to the “boings” of their minke cousins. But much of the ocean’s cacophony comes from significantly smaller creatures – at least 100 invertebrates and many thousands of fish species have been documented making all manner of sounds.

In a world-first global effort, made possible by advances in submersible recording equipment, a team of 17 experts from nine countries has set the goal of recording and collating this underwater orchestra.

Describing their proposed web-based, open-access platform in Frontiers in Ecology and Evolution, the authors say this Global Library of Underwater Biological Sounds (GLUBS) will underpin a new, non-invasive method of monitoring the changing diversity, distribution and abundance of aquatic life, and will even be able to paint a picture of the health of entire ecosystems.

Never had a chance to catch the underwater symphony? Take a listen

Seeking the hidden potential

The creation of comprehensive soundscape recordings has the potential to revolutionise studies of marine life. Many marine species are nocturnal or cryptic, making visual observations difficult or even impossible. By eavesdropping on visually elusive but vocal species, researchers will be able to conceptualise the intricacies of dynamic oceanic environments in greater detail than has been possible with traditional methods.

The library is off to a good start, with existing regional datasets forming the base upon which future recordings will build. According to the authors, there are already millions of recording hours around the world that can be added to the catalogue.

They believe that these existing recordings could be hiding a wealth of as-yet unidentified biological sounds waiting to be assessed. Because most recordings are captured in pursuit of a single target species, little attention has been given to identifying additional sounds.

Collating these recordings in an open-access platform will allow these accidental captures to be investigated and incorporated into studies.

“A global reference library of underwater biological sounds would increase the ability for more researchers in more locations to broaden the number of species assessed within their datasets and to identify sounds they personally do not recognize,” the researchers say.

But the task of matching unknown sounds with known species is immense, and will only grow with every recording added to the library. The researchers hope that they will one day be able to outsource part of this hefty task to artificial intelligence.

Just as AI has enabled facial or voice recognition, and phone apps that identify music and birdsong have become commonplace, AI may one day help scientists distinguish marine life sounds from background noise. As GLUBS expands, it can form the foundation for AI training, which in turn will also facilitate the mining and extraction of marine life sounds from thousands of previously collected recordings.

“Human song varieties include love and work songs, lullabies, chants, and anthems,” says Jesse Ausubel, a founder of the International Quiet Ocean Experiment and a scientist at the Rockefeller University in New York. “Marine animals must sing love songs. Maybe AI applied to the Global Library can help us understand the lyrics of these and many others.”

But the race is on to capture the ocean’s soundscapes before they fall quiet.

All hands on deck to assemble the library

“With biodiversity in decline worldwide and humans relentlessly altering underwater soundscapes, there is a need to document, quantify, and understand the sources of underwater animal sounds before they potentially disappear,” says lead author Dr Miles Parsons, of the Australian Institute of Marine Science.

Without this knowledge, says co-author Aran Mooney, from Wood’s Hole Oceanographic Institution, US, we may lose the chance to characterise healthy systems before they degrade, which could see future restoration efforts flying blind.

“Biodiversity and our ocean ecosystems are in trouble, with healthy coral reefs declining at alarming rates,” he says. “This developing library is a key way to catalogue, monitor and track changes in biodiversity on reefs and other ocean habitats before they are gone [and] also help us define ‘what a healthy reef is’ as we seek to rebuild reefs.”

Recent advances in underwater recording technology will help to accelerate efforts, enabling citizen scientists to contribute to GLUBS.

Low-cost hydrophones and recording systems (such as the HydroMoth) are increasingly available, and waterproof recreational recording systems (such as GoPros) that are popular with snorkellers and divers can also collect underwater biological sounds.

The researchers are excited about this potential.

“Public interest and access to user applications means citizen scientists can drive widespread knowledge sharing,” they say.

“Now is the time to facilitate that progress by gathering the acoustic, ecological, and bioinformatic community together to realize an aquatic-sounds sharing platform.”