In a major scientific breakthrough, artificial intelligence can now assess the risk of dangerous aftershocks within mere seconds of an initial earthquake, according to a new study led by the University of Edinburgh.
The Race Against Time in Seismic Forecasting
Researchers from the University of Edinburgh, the British Geological Survey, and the University of Padua have developed advanced, AI-driven forecasting tools. The team trained sophisticated machine learning models using vast amounts of seismic data gathered from global earthquake hotspots, including California, New Zealand, Italy, Japan, and Greece.
The core objective was to test the AI's ability to generate accurate forecasts predicting how many aftershocks would occur in the 24 hours following a significant earthquake, defined as one with a magnitude of 4 or higher.
Outpacing Traditional Methods
The performance of these new AI models was rigorously compared against the current industry standard, the Epidemic-Type Aftershock Sequence (ETAS) model, which is employed by authorities in Italy, New Zealand, and the United States.
While both forecasting systems demonstrated similar accuracy in assessing aftershock danger, their operational speeds were worlds apart. The established ETAS model required several hours or even days of processing time on a standard mid-range computer to deliver its results. In stark contrast, the new AI models produced a comparable forecast in a matter of seconds.
Implications for Public Safety and Disaster Response
The lead researcher, Foteini Dervisi, a PhD student at the University of Edinburgh’s School of GeoSciences and the British Geological Survey, highlighted the significance of the findings.
"This study shows that machine learning models can produce aftershock forecasts within seconds, showing comparable quality to that of ETAS forecasts," Ms Dervisi stated. "Their speed and low computational cost offer major benefits for operational use: coupled with the near real-time development of machine learning-based high-resolution earthquake catalogues, these models will enhance our ability to monitor and understand seismic crises as they evolve."
This lightning-fast analysis capability promises to be a game-changer for emergency services and government agencies. The near-instantaneous risk assessment provided by the AI-powered tools can significantly improve decision-making regarding critical public safety measures and the strategic allocation of resources in the immediate aftermath of a disaster.
The groundbreaking research, which was supported by the European Union’s Horizon 2020 research and innovation programme, has been published in the scientific journal Earth, Planets and Space.
