Who’s making the hardest beats under the sea these days? We’ll soon find out, because British scientists have trained AI to track the sounds of coral reefs to monitor their health.
It’s not all Little Mermaid and Finding Nemo narratives down there, but regardless, the ocean is a noisy place – especially in areas where sea life is abundant.
In particular, the complex soundscapes discovered amongst coral reefs can offer up valuable insights into the status of their health. More noise around a reef means there’s a bustling community of sea creatures nearby – and that’s a good thing.
But who has time to spend hours or even days listening out for the subtle clicks and crackles of ocean reef-dwellers? Robots, that’s who.
For that reason, scientists based at the University of Exeter have begun using artificial intelligence (AI) to train computers to differentiate between audio recordings of healthy reefs and diminishing reefs.
Though in its early stages, the computers have the ability to detect sounds from a healthy reef with 92 percent accuracy.
Assuming you’re an avid reader on our site – and if you’re not, get with the program – you will likely know that coral reefs are struggling so bad that even Sean Paul is intervening.
In all seriousness, warming waters are causing ocean acidification, destroying entire underwater ecosystems. This leaves octopus, small fish, crustaceans, and sea plant life with no place to thrive and repopulate.
As we inch closer to the limit for global heating, up to 90 percent of coral reefs are expected to be lost by 2040. This includes in sites like the Great Barrier Reef, despite marine activists innovating to protect it.
And although a diver can visually observe what’s happening on a reef, many creatures remain hidden between rocks and anemones, or choose to only emerge at night. This makes it more difficult to achieve certainty about the health status of a reef.
By contrast, hydrophones (underwater microphones) can be left for days, weeks, or even months to record reef activity. Unlike short-length human dives, the recordings provide long-term data sets which tell scientists how the ecosystem is faring.
Leader of the study, Ben Williams, said that tiny shrimps make discreet snapping noises like the ‘crackling of a campfire,’ while other fish make strange grunting, whooping, and knocking sounds.
Hundreds of other sounds occur in the background, easily missed by human ears. AI trained computers are able to detect these though, revealing a more accurate and detailed picture of a reefs’ health status.
The research is being conducted in Indonesia, where the Mars Coral Reef Restoration Project is taking place. Here, hydrophones have been placed alongside coral structures to monitor their growth.
Though the reason fish and other animals are incessantly chattering away is still unknown, catching this activity on audio means a human-nurtured reef is attracting vibrant ocean communities.
Scientists and marine ecologists are excited about sea life monitoring hydrophones. They’re far less costly than hiring diving teams and AI is saving time by analysing recordings at a way faster rate than humans.
They hope to deploy them around the world, like Mexico, The Maldives, and the Great Barrier to monitor reef restoration efforts taking place there.
Deputy Editor & Content Partnership ManagerLondon, UK
I’m Jessica (She/Her). I’m the Deputy Editor & Content Partnership Manager at Thred. Originally from the island of Bermuda, I specialise in writing about ocean health and marine conservation, but you can also find me delving into pop culture, health and wellness, plus sustainability in the beauty and fashion industries. Follow me on Twitter, LinkedIn and drop me some ideas/feedback via email.
Substack is rolling out a new AI detection feature able to scan posts, notes, replies, and comments.
Substack is one of the last safe havens for organic human-written content, right?
This week, the beloved writing platform announced its imminent integration with AI detection software Pangram. It will allow users to instantly scan a post, comment, note, or reply – anything over 100 characters – for the presence of text generated...
Online friendships have long been viewed as less ‘real’ than those formed through face-to-face interaction. However, research indicates that for Gen Z, whose social lives are increasingly shaped by digital spaces, that assumption is outdated.
In a world of 8.3 billion people, how realistic is it to assume that the right people for us are all within a 10km radius?
More than 2,000 years ago, Aristotle defined friendship as a reciprocal...
The intense backlash against digital interventions in sport proves we’re still slow to trust the tech designed to help us.
For all the promises made about Video Assistant Referee (V.A.R) technology, if this World Cup has proved one thing it’s that people don't want perfect decisions. They’d rather have decisions they truly believe in.
V.A.R has, over the course of this summer, become football’s most controversial player.
Recent flashpoints involving Argentina...
A pivot on singledom, the rise of ‘lonely influencers’, and a crackdown on self-optimisation have more and more of us romanticising the lonely life.
I’ve grown up feeling that loneliness is something to hide, or at least feel negative about.
My earliest visualisation of adult loneliness was probably Bridget Jones singing ‘All By Myself’ in her pyjamas, and as I evolved through school, university, and my first jobs, I came to...
We use cookies to ensure that we give you the best experience on our website. If you continue to use this site we will assume that you are happy with it.
0 comments
