"The fault is accumulating stress at a high rate, but this does not suggest that a rupture is imminent," said Yuri Fialko, of the Scripps Institution of Oceanography at La Jolla, California. "When the quake will happen nobody knows."

His study found that, given average annual movement rates in other areas of the fault, there could be enough pent-up energy in the southern end to trigger a cataclysmic jolt of up to 10 metres (33 feet).

"This is new evidence that tells us the same story that we have known for a while," said Scott Brandenberg, a professor at the University of California, Los Angeles school of engineering. "It's a reminder that we need to be ready for it when it happens."

Fialko said said his data taken by satellite is more complete than previous studies because measurements were taken every 20 metres (66 feet) instead of at ground stations 10 kilometers (6.2 miles) apart.

"The observed strain rates confirm that the southern section of the San Andreas fault may be approaching the end of the interseismic phase of the earthquake cycle," he wrote in the science journal Nature.

A sudden lateral movement of seven to 10 metres (23 to 33 feet) would be among the largest ever recorded.

'BIG ONE'

According to the US Geological Survey (USGS), the earthquake that destroyed San Francisco in 1906 was produced by a sudden movement of the northern end of the fault of up to 21 feet (6.4 meters).

Fialko said there had been no recorded movement at the southern end of the fault -- the 800-mile-(1,290 km)-long geological meeting point of the Pacific and the North American tectonic plates -- since the dawn of European settlement in the area.

He said this lack of movement correlated with the predicted gaps between major earthquakes at the southern end of the fault of between 200 and 300 years.

"The longer you wait, the higher the likelihood of rupture," the Scripps scientist said.

Ken Hudnut, a scientist at the USGS, said experts have known since 1988 that the southern section of the San Andreas fault is the most likely source of a "Big One."

"We think recurrence is on a level of a few hundred years and the last one was a few hundred years ago, but we don't understand earthquakes well enough to predict when they will happen," he said.

Fialko's study found that elsewhere on the fault there were average slippage rates up to a couple of centimeters (0.8 inch) a year that prevented the build-up of explosive pressure deep underground.

When these became blocked and then suddenly broke free they produced tremors or earthquakes of varying intensity depending on the movement that had taken place before and the duration of the blockage.