The May 2008 earthquake in Sichuan, China, surprised scientists
Many recent earthquakes may have been the aftershocks of large quakes that occurred hundreds of years ago, according to scientists.
In the journal Nature, researchers described a new pattern in the frequency of aftershocks that could explain some major quakes.
They found that, away from plate boundaries, echoes of past earthquakes can continue for several hundred years.
Here, in the middle of a continent, the earth takes longer to recover.
It's something we had never spotted before," said Seth Stein from Northwestern University in Illinois, US.
"Most big earthquakes happen at [plate] boundaries - like the San Andreas fault. There is a lot of movement there and aftershocks go on for about ten years after a big quake."
When the aftershocks have dissipated, scientists monitor regular movement of the earth to gauge the likelihood of a future quake.
The researchers gathered data from faults in the US
But small earthquakes also occur where there is none of this regular movement, he explained. "So if the ground has not been storing up energy for future earthquakes, these must be aftershocks."
This, the scientists say, could explain the disastrous earthquake in 2008 in China's Sichuan province. The event shocked many scientists as this was an area where there had been hardly any earthquakes in the past few centuries.
But these "aftershock quakes", the scientists say, get smaller over time.
"It even looks like we see small earthquakes today in the area along Canada's Saint Lawrence valley where a large earthquake occurred in 1663," Professor Stein said.
"If you look at where they are - they're on the fault plane of the big earthquake."
He and his colleague, Mian Liu from the University of Missouri, found the same pattern repeated in seismic data from faults around the world.
This discovery could help scientists to foresee the location of big earthquakes.
"Predicting big quakes based on small quakes is like the 'whack-a-mole' game," Professor Stein explained. "You wait for the mole to come up where it went down.
"But we now know the big earthquakes can pop up somewhere else."
He recommended that, instead of just focusing on the regions where small, regular earthquakes happen, scientists should use methods like GPS satellites and computer modelling to look for places where the earth is "storing up energy for a large future earthquake".
Tom Parsons, a scientist from the US Geological Survey (USGS) in California was not involved in this study, but wrote an accompanying article in the same issue of Nature, explaining its significance.
He said that with a more comprehensive approach to studying earthquakes, researchers would eventually be able to "arrive at a practical solution" - balancing the available resources with the need to protect areas that were at risk.