Artificial Intelligence

AI tunes in to rescue coral reef

26th September 2024
Sheryl Miles
0

Coral reefs are critical to our planet’s ecosystem. They provide food, shelter, and protection for marine life, and they act as natural barriers, protecting coastlines from storms and erosion.

They also support the livelihoods of more than half a billion people through food, income, protection, and tourism. Looking beyond coastal areas, they also play a key role in marine biodiversity.

However, these fragile ecosystems are increasingly threatened by climate change, pollution, and overfishing. So, as coral reefs face unprecedented degradation, the need for solutions to monitor and protect them is growing.

The sound of a healthy reef

A surprising yet effective method of monitoring coral reef health comes through sound. Healthy coral reefs are vibrant ecosystems that teem with life, producing a distinctive “crackling” noise which is the culmination of a chorus of snapping shrimp, fish calls, and other marine sounds. In contrast, degraded reefs are much quieter, lacking the bustling marine life that characterises a healthy reef.

Scientists have long known that sound can provide important clues about the state of a reef, but manual assessments can be time-consuming and costly.

This is where ReefSound, a new AI-powered solution developed by three Eton College students, Simba Shi, Thenuk Rodrigo, and Bonnie Li, comes into play.

The birth of ReefSound

The idea for ReefSound was born out of a conversation between 17-year-old Rodrigo and Shi. After reading about coral reefs and watching documentaries that highlighted the scale of their degradation, the two realised the importance of developing a more efficient and scalable way to assess reef health. Their solution came in the form of sound, inspired by the fact that healthy and degraded reefs emit different acoustic signatures.

The duo, joined by their fellow student Bonnie Liu, set out to develop an AI analysis tool that could accurately assess coral health using just a short recording of underwater sounds.

By analysing a 60-second sound byte, ReefSound’s AI model can determine the health of a coral reef with 99.4% accuracy. This breakthrough is a leap forward in reef monitoring, as it is both faster and cheaper than traditional methods like manual diver surveys or satellite imaging.

The technology

At the core of ReefSound is an AI model that has been trained to recognise the acoustic differences between healthy and degraded coral reefs. The model uses machine learning algorithms to analyse sound recordings and identify patterns that indicate the presence or absence of marine life.

According to the students, the system is 60% cheaper and five times more efficient than current reef assessment techniques. Moreover, it can be deployed remotely, providing a scalable solution for conservation organisations around the world.

ReefSound’s potential has not gone unnoticed. The startup caught the attention of Nasa and is currently in discussions with coral preservation organisations in seven countries, including Malaysia, the US, and Australia. Its technology has been used to monitor coral reefs off the coast of Cozumel, Mexico, and the students are hopeful that it will be adopted on a larger scale by governments and conservation agencies worldwide.

The future of ReefSound

Despite balancing their A-level studies in maths, further maths, physics, and computer science, Shi, Rodrigo, and Li have ambitious plans for ReefSound. Their vision is to see the technology used in every coral reef around the world, helping conservation efforts by providing fast, accurate, and affordable assessments of reef health.

In addition to securing scholarships worth $298,000 (£227,000) from US universities, the students have won prestigious accolades for their work, including becoming the first British team to win the Conrad Challenge, a high school STEM competition run by Space Center Houston.

Looking ahead, the ReefSound team aims to continue refining their technology and expanding its reach. They are hopeful that their AI model will be adopted by governments for large-scale reef monitoring projects, helping to protect coral reefs from further degradation. For the three young innovators, the project is more than just a technical achievement, it is a passion-driven mission to make a tangible impact on the fight against climate change and marine conservation.

Featured products

Product Spotlight

Upcoming Events

View all events
Newsletter
Latest global electronics news
© Copyright 2024 Electronic Specifier