New AI system can predict earthquakes

Rodiano Bonacci
Ottobre 24, 2017

"This is the first time that machine learning has been used to analyse acoustic data to predict when an natural disaster will occur, long before it does, so that plenty of warning time can be given - it's incredible what machine learning can do", said Colin Humphreys, professor at Cambridge's Department of Materials Science and Metallurgy.

A group of analysts from the U.S. and United Kingdom identified a hidden signal emitted before a tremor, then used that "fingerprint" to train an algorithm to anticipate seismic activity.

Scientists have developed an artificial intelligence (AI) system to successfully predict earthquakes, an advance that may help prepare for natural disasters and potentially save lives. But despite advances in instrument precision, geoscientists still struggle to determine timing and severity.

The researchers from the University of Cambridge, Los Alamos National Laboratory and Boston University, identified a hidden signal leading up to earthquakes, and used this "fingerprint" to train a machine learning algorithm to predict the timing, but not the size, of future earthquakes.

The characteristics of this sound pattern can be used to give a precise estimate of the stress on the fault and to estimate the time remaining before failure, which gets more and more precise as failure approaches, according to the study.

Using a lab-based system that mimics real earthquakes, the researchers applied machine learning techniques to analyse the acoustic signals coming from the "fault" - on which earthquakes occur.

Not yet ready for primetime, the laboratory experiment (which really ought to have some kind of punny name) could eventually be scaled up for use at sites-like California's San Andreas Fault or the Cascadia fault in the Pacific Northwest, where small repeating earthquakes are similar to those tested in the lab.

Machine learning enables the analysis of datasets too large to handle manually and looks at data in an unbiased way that enables discoveries to be made, researchers said.

Let us know what you like about Geek by taking our survey.

Altre relazioniGrafFiotech

Discuti questo articolo

Segui i nostri GIORNALE