Scientists have developed a method to assess whether tiny changes in Earth's gravity can be used to predict oncoming earthquakes. Their conclusion: It can't, which refutes a hypothesis which researchers thought would provide hope for better warnings.
As one of the most mysterious of our planet's natural disasters, earthquakes' location and timing have been impossible to precisely predict. But scientists do know that the most destructive of these events occur in subduction zones, regions where tectonic plates often collide and shift past one another.
These zones of immense activity make them ideal places to search for phenomena that could herald future quakes. To that end, previous studies have suggested that satellite data could be the key to early warnings months ahead of big earthquakes, challenging the belief that large earthquakes are immune to precision forecasting in time.
Now, by analyzing measurements made by NASA's GRACE and GRACE-FO satellites, researchers found that relying on precursors formed from single data points is largely invalid, said Lei Wang, author of the study and an associate professor in civil, environmental and geodetic engineering at The Ohio State University.
"Earthquake risk can be affected by a lot of different factors," said Wang. "Many also happen over hundreds and tens of hundreds of years apart, so a few decades of modern data isn't sufficient enough to accurately predict them." The research was recently presented at the annual meeting of the American Geophysical Union.
While scientists have gained insights into the processes that trigger earthquakes, being able to offer advance notice of these volatile events is still a challenge they haven't overcome. Instead, such programs work by detecting fast-moving seismic waves quickly enough that those nearby can be alerted, usually only seconds to minutes before shaking arrives.
In contrast to earthquake data gathered by sensor instruments on the ground, the twin GRACE satellites monitor changes in Earth's gravity and track mass water redistribution in reservoirs over land, ice and in the oceans. In this study, the team used their data to determine if large-scale gravity changes had been detected several hundred miles below ground in the months before the 2010 8.8 Maule earthquake in Chile and the 2011 9.0 Tohoku megathrust earthquake in Japan.
After comprehensively analyzing and comparing multiple gravity data solutions with other anomalous GPS statistics around the globe, the team concluded that satellites are no better at predicting changes under the earth than any other conventional geodetic techniques.
"If satellites actually could detect anomalous changes to Earth's uniform gravity, it would revolutionize our understanding of earthquake mechanisms and shed new light on prediction and mitigation science," said Wang. "But there's no solid evidence for it."
Even if satellites could reveal signs of future earthquakes, it's likely this method wouldn't be useful on a global scale, especially in areas that show distinct earthquake mechanisms, said Dhamsith Weerasinghe, co-author of the study and a PhD student in civil, environmental and geodetic engineering at Ohio State.
"Even if we modeled a known place, we couldn't apply those findings to another because of the different geometry and different material in that environment," said Weerasinghe. "It'd take too long to discover and understand certain changes."
To keep refining the method, the team plans to examine if there were any notable gravity precursors to the 8.8 earthquake that took place this year in Kamchatka, Russia. In the meantime, Wang hopes this work will help future scientists learn how to combine historical data with new advances in geodesy and environmental monitoring.
"To accurately predict an earthquake that is days or hours away, the technology we have now is a little premature," said Wang. "That's why we need some young people with passion to jump in and make promising strides in the field."
