For a brief period in 2011, just after two tectonic plates gave way off the eastern coast of Japan, gravity wobbled. The Earth’s gravitational field is the result of a distribution of matter—a slightly firmer tug where the world is denser; a looser grasp where it is not. When massive volumes of earth and water are suddenly displaced, like in an earthquake, that distribution changes. The forces that hold the moon close, keep the atmosphere thick, and tie our feet to the ground jerked into a new alignment. The whole world tripped, seconds before the seismic waves arrived.
Not that anyone noticed. Even the very biggest tremors, like the 2011 earthquake, have a subtle effect on gravity. But for seismologists accustomed to listening closely to the Earth’s rumblings, such changes have long offered a tantalizing possibility: an earthquake signal that is practically instantaneous, spreading through the globe at the speed of light. In recent years, scientists have scoured data from big quakes for signs of these gravity perturbations. They’re elusive, and still fairly controversial in seismology. But with the help of more sensitive instruments and better computer models, the hunters have started to find them.
Now they’re getting closer to putting that data to use. In a paper published in Nature, researchers describe an earthquake early warning system that relies on those gravity-derived signals alone. They tested their model on seismic data from the earthquake, finding it could accurately detect the quake about eight seconds faster than previous methods and give a better estimate of its massive size. The work is a proof of concept, looking back on a single event. But it’s meant to test if the method could add precious seconds to early warning systems in the future. We are showing that this is actually a signal, people weren’t even looking at this part of the data, but it is comparable, if not better than, existing signals.
Those existing signals are primarily P-waves, seismic ripples that occur as rock compresses and vibrates from a sudden shock. When these waves reach seismic stations, software quickly pinpoints where the earthquake originated and estimates its size. The goal is to give people a heads-up, however brief, before the up-and-down motion of S-waves, a slower type of tremor that often causes the most damage. In recent years, better instruments and algorithms have resulted in faster and more reliable warning systems. But P-waves typically only travel at a few kilometers per second, putting a theoretical limit on the speed of detection.
Gravity perturbations are quicker. But they’re also far less forceful than P-waves, making them tricky to pick out from seismologists’ greatest enemy: noise. The din of the earth is constant, a chorus of tiny events generated by people, seismic tremors, and air and ocean turbulence that makes the early hints of a major quake exceedingly difficult to hear. Seismologists want a clear signal of what’s coming.
For decades, seismologists have debated whether a clear detection is possible. There are tools to observe gravitational waves directly, like the massive LIGO facilities in Louisiana and Washington. But they’re mostly of use to astronomers and aren’t practical for picking up the tiny shifts caused by earthquakes. Instead, the fluctuations are observed indirectly by seismometers, which pick up the response of the Earth as it pushes and pulls away to counteract the shift in mass. Trouble is, the gravity changes and the elastic responses to them mostly cancel each other out.
Seismic waves from a big quake are easy to see. For example, think of the classic image of a seismograph, pencil scratching out telltale waves on a rotating paper as the tremor arrives. Even to highly trained eyes, PEGS are just squiggles, indistinguishable from the noise. It’s hard to prove they’re there. In 2017, early identification f PEGS in seismic data received from other seismologists.
You must be logged in to post a comment.