To locate the fault the researchers used ground penetrating radar, seismic reflections and electrical resistance measurements. Then they dug trenches along the fault and collected soil samples in which they analyzed the age and location of carbon 14 in organic material. Because one side of the fault dropped each time there was an earthquake along the fault, soil would wash over from the higher side to the lower side.

By dating the age of organic matter in the soil layers, the group was able to determine when the shifts had occurred and determine the frequency of the events. In the past 8500 years, the movement along the fault has lifted one side 1.8 meters (2 yards) higher than the other.

Based on their lab work, the researchers concluded that the most recent earthquake occurred between 610 and 1475 AD and that there had been two more, one between 850 BC and 850 AD and one between 850 BC and 6480 BC. The estimate is consistent with a 6 to 6.5 Richter scale earthquake in 1356 in Basel that destroyed many buildings.

Based on frequency of the past earth movements, the scientists think that the recurrences will happen every 1500 to 2500 years, making the next big Basel temblor due between 2856 and 3856. Their estimate is based on the period of time that has elapsed between previous quakes on the fault.

One of the lead researchers, Domenico Giardini, told United Press International, "In general, in Europe, we never find the faults that produce earthquakes. It's not like in California where you see the San Andreas. We have a lot a faults in Europe but usually they're not active." Because a lot of the faults have not been identified, many of the scenarios for earthquake occurrence have been vague and imprecise, Giardini said.

"It is really a relatively bad scenario for the city, which in a sense was known, because we know that the city was destroyed in 1356. But now we know why and we are able, in all the different parts of the city forecast, to know what the shaking will be in the next event more precisely than we could have done before," Giardini said in a telephone interview from Zurich. Giardini is director of the Swiss Seismological Service and a professor of seismology at the Swiss Federal Institute of Technology.

Paul Bodin, a seismologist at Center for Earthquake Research and Information at the University of Memphis said he agreed with the conclusions of the paper. Bodin said that he was impressed that the researchers were able to locate the fault. "The fault is not long. Eight kilometers (5 miles) is a short fault," Bodin said. "The work looks solid and the inferences they've drawn look reasonable," he said. Bodin is studying the mechanics of faulting and seismic rupture, with application to seismic hazard assessment

Jonathon Stewart, assistant professor of civil engineering University of California at Los Angeles, told UPI that if the research had found that there had been very little movement over the last 10,000 years it would mean it was a less active region and that not as many seismic precautions would need to be taken. "If you have a region that's very seismically active, then you're going to design buildings to more stringent standards, than if earthquakes happen relatively less often," Stewart said.

Speaking of the 1500 to 2500-year interval the researchers predict, Stewart told UPI, "1500 years on a geologic time scale is pretty frequent, actually. . . . In California we define a fault as active if it's ruptured in the last 10,000 years." Stewart specializes in geotechnical earthquake engineering and foundation engineering.

Researchers from the University of Basel, the University of Strasbourg and Swiss Federal Institute of Technology teamed up to complete the study, which is reported in this week's issue of the journal Science.

(Reported by Joe Grossman in Santa Cruz, California)

Copyright 2001 by United Press International.