EVANSTON, III — Data scientists and seismologists may have discovered a way to predict what they previously thought was unpredictable: earthquakes.
The interdisciplinary team from Northwestern University found that so-called “slow” earthquakes potentially lead to “regular” earthquakes. Slow earthquakes release energy at a gradual rate, stretching from hours to several months. What we consider to be regular earthquakes release energy much faster and suddenly.
The researchers’ findings could help seismologists and geologists to narrow down a time frame when a major earthquake could occur.
“While the build-up of stress in the Earth’s crust is largely predictable, stress release via regular earthquakes is more chaotic in nature, which makes it challenging to predict when they might occur,” explains data science scholar Kevin Chao, who works in the Northwestern Institute on Complex Systems and is the study’s first author. “But in recent years, more and more research has found that large earthquakes in subduction zones are often preceded by foreshocks and slow earthquakes.”
The researchers examined a region in Taiwan that has over 100 seismic stations that have continuously recorded ground motion for years. They found deep tremors there, slow earthquakes that recur in cycles days or weeks long. To detect and monitor these deep tremors, Chao and his team created highly sophisticated algorithms and applied them to data collected from ten of the seismic stations in the area. They noticed deviations in the deep tremors’ behavior in the weeks leading up to Taiwan’s major earthquake in March 2010.
“After the 6.4-magnitude earthquake occurred, we noticed a potential to study deep tremor near the event,” says Chao. “We identified the increase in tremor duration three weeks before the earthquake, but we initially could not draw conclusions because tremor rates increase all the time and for different reasons.”
But three years later, by comparing data from other seismic station and earth observatories, like global positioning satellites, with their their statistics and algorithms, the team showed that deep tremor pattern changes could signal an impending earthquake nearby. To be sure, the authors looked at four other earthquakes and found similar preexisting conditions and patterns ahead of the quakes.
They hope the findings will lead to more interest in their theory and perhaps to better quake predictions in the future.
“Much more data analysis of these tiny but fascinating tremor signals is necessary,” says Chao, “before mid- to short-term earthquake forecasting become reliable.”
The full study was published June 27, 2017 in the Journal of Geophysical Research: Solid Earth.