Researchers find satellite data can’t forecast future tremors — here’s what that means
But here’s the controversial takeaway: even cutting-edge gravity measurements from space aren’t offering the earthquake warning boost scientists hoped for.
Columbus, Ohio — A team of researchers has tested whether tiny shifts in Earth’s gravity could signal an upcoming earthquake. Their verdict: such precursors don’t reliably predict quakes, challenging a line of thinking that held promise for longer, more actionable warnings.
Earthquakes remain among the most enigmatic natural disasters, with location and timing notoriously hard to forecast. We do know that the most devastating events often strike along subduction zones, areas where tectonic plates collide and slide past each other.
These highly active regions were prime grounds for exploring potential early signals. Earlier work suggested satellite observations might reveal clues months before major earthquakes, casting doubt on the long-held belief that large quakes could be forecast with precision.
In the new study, researchers analyzed data from NASA’s GRACE and GRACE-FO satellites to see if a single data-point precursor could foretell future earthquakes. Lei Wang, an associate professor of civil, environmental, and geodetic engineering at The Ohio State University and the study’s author, explains that the risk of earthquakes depends on many factors. He notes that many quakes occur with centuries between them, so a few decades of modern data aren’t enough to yield reliable predictions.
The findings were shared at the annual meeting of the American Geophysical Union.
Traditionally, forecasting advances in seismology have focused on detecting rapidly propagating seismic waves and issuing alerts to nearby populations seconds to minutes before shaking arrives. By contrast, satellite measurements monitor variations in Earth’s gravity and track how water mass moves across land, ice, and oceans. The researchers compared gravity-change signals from the Maule earthquake in Chile (2010, magnitude 8.8) and the Tohoku earthquake in Japan (2011, magnitude 9.0) with various global gravity and GPS data to test whether large, pre-earthquake gravity anomalies could be detected hundreds of miles underground in the months leading up to those events.
The team’s conclusion is clear: gravity data from satellites do not outperform traditional geodetic methods in identifying underground changes before quakes.
“If satellites could detect anomalous gravity changes before earthquakes, it would revolutionize our understanding of quake mechanisms and transform prediction and mitigation strategies,” Wang says. “But there isn’t solid evidence to support that possibility.”
Even if such signs existed, translating them into a global forecasting tool would be difficult. Co-author Dhamsith Weerasinghe, a PhD student at Ohio State, emphasizes that what works for one region might not apply elsewhere due to differences in geology, geometry, and materials. It would take extensive, location-specific work to interpret any potential signals reliably.
Looking ahead, the researchers plan to revisit gravity data for other significant events, such as the 2023 Kamchatka quake, to see if any notable precursors emerge. In the meantime, Wang stresses that the current technology is not yet ready to deliver day- or hour-scale warnings, but hopes the work will help guide future efforts to combine historical datasets with advances in geodesy and environmental monitoring.
“Right now, we’re not at the point where we can predict earthquakes with days or hours of lead time,” he says. “That’s why we need motivated researchers to push the boundaries and explore promising approaches.”
—
Contact: Lei Wang, Wang.1115@osu.edu
Written by: Tatyana Woodall, Woodall.52@osu.edu
Would you like this rewritten piece tailored for a specific audience (e.g., students, policymakers, or the general public) or adjusted for a particular publication style (more casual, more formal, or more technical)?
