Massive earthquakes don’t just shake the earth – they make speed of light adjustments to the Earth’s gravitational field. Researchers have now trained computers to identify these tiny gravitational signals and to mark the location and size of a strong quake almost instantly.
It is a first step to creating an early warning system for the planet’s most powerful quakes, say scientists in the journal Nature.
Current seismic wave based detection methods have difficulty distinguishing between, for example, a magnitude 7.5 and magnitude 9 quake in the few seconds following such an event.
However, seismic waves are not the earliest signs of a quake. All the mass moving around in a big earthquake also changes the density of the rocks at different locations. Those shifts in density equate to tiny changes in Earth’s gravitational field, producing ‘elastogravity’ waves that travel through the ground at the speed of light.
Andrea Licciardi, a geophysicist at the Université Côte d’Azur in Nice, France and his colleagues created PEGSNet, a machine learning network designed to identify ‘“Prompt ElastoGravity Signals.” The researchers trained their machines on a combination of real seismic data collected in Japan and 500,000 simulated gravity signals for earthquakes in the same region. The synthetic gravity data are essential for the training, says Licciardi, because the real data are so scarce.
The algorithm was trained to look for a single point for an earthquake’s origin. But closer up the origin is actually a larger region that has ruptured. If scientists want an accurate estimate of where a rupture occurred in the future, the machines need to look for regions, not points, adds Vallee.
www.sciencenews.org/article/machine-learning-gravity-earthquake-ai