Scientific experiments cannot predict the exact timing or location of upcoming earthquakes, but a decade-long evaluation of 27 forecasting models reveals that specific systems can reliably estimate patterns in seismic activity. The findings offer new evidence to help agencies better communicate seismic risks following tests against hundreds of real-world events.
California Testing Evaluates 27 Models Against 597 Earthquakes
Researchers evaluated 27 earthquake forecasting models using more than 51,000 forecasts generated between August 2007 and August 2018. The international scientific collaboration through the Collaboratory for the Study of Earthquake Predictability tested whether the models accurately described the number, locations, and magnitudes of seismic events.
The study compared these predictions with 597 recorded earthquakes of magnitude 3.95 or greater within California. Because the forecasting systems generated next-day predictions before the earthquakes occurred, researchers assessed performance against new observations rather than historical data. The study was published in Nature Communications in August and highlighted in a University of Bristol news release.
We’ve provided a model ‘recipe’ identifying elements that can be considered when developing forecasts consistent with the observed number, spatial locations and magnitude distribution of earthquakes,
said lead author José A. Bayona, Early Career Leverhulme Research Fellow in the University of Bristol’s School of Earth Sciences.
Statistical Consistency Shows Varying Success Across Models
The accuracy of the models depended heavily on which aspect of earthquake activity the researchers examined. Approximately 95% of the models produced forecasts statistically consistent with the observed number of earthquakes. About two-thirds matched the locations of recorded earthquakes, while roughly 95% matched the distribution of earthquake magnitudes.
These figures describe how many models passed statistical consistency tests rather than indicating that the models predicted 95% of earthquakes correctly. Earthquake locations proved harder for the systems to describe consistently compared to earthquake counts or magnitude distributions. The U.S. Geological Survey notes that no scientific method currently predicts the exact time, location, and magnitude of an impending earthquake with reliable accuracy, though agencies can calculate probabilities within specific areas and time frames.
High Stress Levels on Fault Systems Raise Regional Concerns
California continues to face long-term risks along major faults like the southern San Andreas Fault, where the last major rupture occurred in 1857 during the magnitude 7.9 Fort Tejon earthquake. A 2026 study from researchers at the University of Hawaiʻi at Mānoa found that stress levels on the southern San Andreas and San Jacinto fault systems sit at their highest levels in roughly 1,000 years.
Right now, with stress at historically high levels across the region and more than 160 years elapsed since the last major rupture, the system is in a critically loaded state,
said study lead author Liliane Burkhard. Researchers emphasize that while stress levels are high, this data does not mean a major earthquake is imminent or provide a countdown clock for the next seismic event.
Common Questions About Seismic Forecasting
Can scientists predict when and where an earthquake will strike?
No scientific method can currently predict the exact time, location, and magnitude of an impending earthquake. According to the U.S. Geological Survey, scientists can only calculate the probability of earthquakes occurring within a particular area and time frame.
What did the decade-long California experiment actually prove?
The study evaluated 51,625 next-day forecasts generated between 2007 and 2018 against 597 recorded earthquakes in California. The experiment demonstrated how selected models can consistently estimate patterns in earthquake activity when tested against future observations.
Do earthquake swarms mean the Big One is coming?
A stretch of heightened activity or an earthquake swarm does not give scientists a reliable way to determine whether a larger earthquake is about to occur. Hundreds of small earthquakes do not amount to a countdown clock for a major rupture.
Keep reading