Two minutes after the world’s largest tectonic plate shook off Japan’s coast, the weather agency issued a final warning to some 50 million inhabitants: An 8.1-magnitude earthquake triggered a tsunami that rushed to the coast. But experts did not assess the true magnitude of the March 11, 2011 Tohoku earthquake until hours after the waves arrived. It was eventually rated a 9 – releasing more than 22 times the energy experts predicted, killing at least 18,000 people, and some affected areas never being alerted. Now scientists have found a way to estimate the magnitude of earthquakes faster and more accurately by using computer algorithms to identify the wakes of gravitational waves that shoot off faults at the speed of light. Researchers involved in the search for gravitational waves (ripples in space-time created by the motion of massive objects) realized that these gravitational signals, traveling at the speed of light, could also be used to monitor earthquakes. “The idea is that as long as the mass moves anywhere, the gravitational field changes, and … everything feels it,” said Bernard Whiting, a physicist at the University of Florida who works at the Laser Interferometer Gravitational-Wave Observatory. Surprisingly, these signals even show up in seismometers.” In 2016, Whiting and colleagues reported that conventional seismometers could detect these gravitational signals. Earthquakes cause massive changes in mass; these changes create gravitational effects that deform both the existing gravitational field and the ground beneath the seismograph. By measuring the difference between the two, scientists concluded that they could create a new kind of earthquake early warning system. Gravitational signals appear on seismographs before the first seismic waves arrive, and this part of the seismogram has traditionally been ignored. By stacking the signals from dozens of seismometers on top of each other, scientists can identify patterns that explain the size and location of large earthquakes, Whiting said. Now postdoc Andrea Licciardi at the Cote d’Azur University in France and colleagues have built a machine learning algorithm for pattern recognition. They trained the model with hundreds of thousands of simulated earthquakes, then tested it with a real dataset of the Tohoku earthquake. The model accurately predicted the magnitude of the quake in about 50 seconds — faster than other state-of-the-art early warning systems, the researchers report in the journal Nature. “It’s not just a seed of an idea — they’re showing that it can be done,” Whiting said. “What we’re showing is a proof-of-principle. What they’re showing is a proof-of-principle.”
This article is reprinted from: https://www.solidot.org/story?sid=71517
This site is for inclusion only, and the copyright belongs to the original author.