What happens in the moments before a natural disaster can mean the difference between life and death. Now, scientists at Cardiff University have developed a warning system that gives people in the road time to escape.
To Sophia Guimer, News Reporter
Tuesday 25 April 2023 17:14 UK
Scientists have developed an early warning system for tsunamis and underwater earthquakes using artificial intelligence.
A team at Cardiff University has successfully combined AI with techniques such as underwater microphones to monitor the movement of tectonic plates in real time.
Experts are already using recordings to detect 200 earthquakes in the Pacific and Indian oceans.
Dr Usama Kadri, co-author of the study and Senior Lecturer in Applied Mathematics at Cardiff University, said:
“[They] It travels through water much faster than a tsunami, giving it more time to evacuate a location before making landfall. ”
Acoustic gravitational waves are sound waves that travel at the speed of sound in the deep ocean. They can be generated by underwater earthquakes, explosions, and landslides, all of which can cause tsunamis.
The new system uses underwater microphones, or hydraphones, recordings and “computational models” to triangulate the sources of tectonic movements thousands of kilometers away.
Mr. Cuddy and his partner Dr. Bernabe Gómez Pérez are currently installing 11 hydrophones in oceans around the world. He said that 24 would be enough to monitor the entire planet.
read more:
German publisher apologizes for AI-generated interview with Michael Schumacher’s family
Google Boss Admits AI Dangers ‘I Can’t Sleep At Night’
Existing warning systems rely on waves reaching ocean buoys and seismic sensors to trigger tsunami warnings.
This leaves little time for evacuation and can take a few extra minutes in potentially life-or-death situations.
However, they are not always accurate in predicting the hazards posed by the resulting tsunami.
Kadri told Sky News:
“In the case of the Sri Lanka tsunami [Indian Ocean] In 2004, we could have given people 65 minutes to evacuate. It has the potential to save almost any life. ”
“When the wave hits the hydrophone, it takes about 17 seconds to calculate.
He explained that their system was designed to work in tandem with existing systems, with both acting as checks and balances for the other.
Algorithms in the new system can classify earthquake “slip types” and magnitudes before analyzing earthquake characteristics such as length and width, uplift rate and duration.
This allows us to know the actual size of the tsunami.
Co-author Gomez Perez, Ph.D., now at the University of California, Los Angeles, said, “Trust movements with strong vertical slip elements are more likely to raise or lower the water column compared to horizontal slip elements.
“Thus, knowing the type of slip early in the evaluation can reduce false alarms and complement and enhance the reliability of warning systems through independent cross-validation.”
Cuddy said false alarms could lead to closures of businesses and ports, which could have major financial implications.
“People lose faith in the system when you’re crying wolf, and when it becomes a reality, people don’t leave,” he added.
The team is meeting with the UNESCO Marine Commission to discuss opportunities to use the technology for disaster prevention. Portugal, famous for its huge waves, were particularly interested in the insight it could offer.
The team’s work to predict tsunami risk is part of a long-term project to strengthen natural disaster warning systems around the world.
The full report was published in Physics of Fluid on April 25, 2023.
