A new tool that assesses the level of danger posed by tsunamis in real-time has been made operational on a global scale.
The early warning system, Global Real-time Early Assessment of Tsunamis (GREAT), uses sound waves that travel through the ocean much faster than tsunamis and can be detected by underwater microphones called hydrophones.
Led by a team from Cardiff University, the technology could help save lives by giving people more time to act and evacuate, and help reduce false alarms, which damage trust in warning systems.
The technology, presented in Geoscientific Model Development , accessed Comprehensive Nuclear-Test-Ban Treaty Organization (CTBTO) real-time hydroacoustic data via the Instituto Português do Mar e da Atmosfera (IPMA) since June 2024, for live operational testing at the newly established Tsunami Warning Centre in Cardiff.
Project lead Dr Usama Kadri, a Reader in Applied Mathematics at Cardiff University, said: "Our work shows that we can use sound waves in the ocean to assess tsunamis quickly and globally, as they are happening and not after the wave has already hit land.
"The technology runs in real-time, without relying on pre-set templates or assumptions, and has been tested successfully using real hydrophone data.
"This next phase of testing via IPMA, assesses the system's performance under actual operating conditions, addressing challenges such as hardware limitations, data transmission delays, and potential sensor failures."
All of this will make it possible to issue earlier and more reliable warnings that could potentially save lives in at-risk communities across the globe.
Existing warning systems are often based on sea-level sensors like DART buoys that only detect tsunamis after they arrive, leaving little time to respond.
By the time the signal is clear to warning centres, it may already be too late, especially for areas close to the source like the Sumatra tsunami in 2004, when the first waves arrived just 14 minutes after the earthquake.
"Seismometers are also employed to alert warning centres when an earthquake occurs, but they are unable to determine the size of the tsunami if one is generated," says Dr Kadri.
Instead, GREAT can complement existing warning infrastructures, working alongside them for independent validations powered by analytical models and advanced machine learning, to detect faster-travelling sound waves, generated at the same time as the tsunami, to get an earlier and more accurate picture of what is coming.
Traditional warning approaches are also incapable of assessing tsunamis generated from non-seismic sources such as landslides, volcanoes, meteotsunamis, meteorites and explosions, among others.
GREAT can capture the properties of tsunamis from these sources because they also generate acoustic gravity waves.
However, a major challenge for the emerging technology is the limited number of hydroacoustic stations, the team admits.
The CTBTO operates 11 stations globally, but only four hydrophone stations currently provide real-time data via IPMA.
Their sparse geographic distribution further limits regional coverage.
Dr Kadri added: "Provided an event is within a thousand kilometres from a hydrophone station, it will take on average, six minutes for an end-to-end assessment.
"The analysis itself takes only a few seconds on a standard PC station.
"To enable global early warning, approximately two dozen hydrophone stations would be needed - a relatively modest number compared to other existing monitoring systems."
GREAT has been endorsed by the United Nations as an Ocean Decade project .
Dr Kadri presented the technology at the First Ocean Decade International Coastal Cities Conference: Better Ocean, Better City in Qingdao, China, and the Thirty-first Session of the Intergovernmental Coordination Group for the Pacific Tsunami Warning and Mitigation System in Beijing, China.
The paper, 'GREAT v1.0: Global Real-time Early Assessment of Tsunamis', is published in Geoscientific Model Development .
