Researchers have created an AI-based tsunami warning system. 

Scientists at the University of California, Los Angeles and Cardiff University have developed a new early warning system that uses acoustic technology and artificial intelligence (AI) to predict the risk of tsunamis. 

Writing in Physics of Fluids, the researchers explain that determining the type of earthquake that triggers a tsunami is critical in assessing the level of risk. 

To do this, the scientists propose using underwater microphones, known as hydrophones, to monitor tectonic activity in real-time. 

The researchers also used AI algorithms to classify the slip type and magnitude of the earthquake, allowing them to calculate important properties such as the effective length and width of the wave, the uplift speed, and duration, which can dictate the size of the tsunami. 

The system was tested using available hydrophone data and was found to be almost instantaneously effective in describing the earthquake parameters with low computational demand. 

The scientists plan to improve the system by adding more information to increase the accuracy of tsunami characterisation. 

The system could be useful in enhancing the safety of offshore platforms and ships as part of a larger project to improve hazard warning systems. 

Co-author Bernabe Gomez said that “knowing the slip type at the early stages of the assessment can reduce false alarms and enhance the reliability of the warning systems through independent cross-validation”.