By
updated 10/13/2014 5:15:38 PM ET 2014-10-13T21:15:38

With several faults slicing through the San Francisco Bay Area, forecasting the next deadly earthquake becomes a question of when and where, not if.

Now researchers propose that four faults have built up enough seismic strain (stored energy) to unleash destructive earthquakes, according to a study published today (Oct. 13) in the Bulletin of the Seismological Society of America.

The quartet includes the Hayward Fault, the Rodgers Creek Fault, the Green Valley Fault and the Calaveras Fault. While all are smaller pieces of California's San Andreas Fault system, which is more than 800 miles (1,300 kilometers) long, the four faults are a serious threat because they directly underlie cities. [ Photo Journal: The Gorgeous San Andreas Fault ]

"The Hayward Fault is just right in the heart of where people live, and the most buildings and the most infrastructure," said Jim Lienkaemper, lead study author and a research geophysicist at the U.S. Geological Survey's Earthquake Science Center in Menlo Park, California. "But it's not just one fault, it's the whole shopping basket. If you are in the middle of the Bay Area, you are near a whole lot of faults, and I'm concerned about all of them."

Lienkaemper and his colleagues gauged the potential for destructive earthquakes by monitoring tiny surface shifts along California faults. Certain faults are in constant motion, creeping steadily by less than 0.4 inches (1 centimeter) each year. These slow movements add up over time, cracking sidewalk curbs and buildings. They also serve as clues to what's happening deep below ground, where earthquakes strike.

"If you figure out where faults are creeping, it tells you where they're locked and how much they're locked," Lienkaemper told Live Science.

Fault creep varies, with some faults sliding at a snail's pace and others barely budging. Models suggest that the diversity comes from locked zones that are 3 to 6 miles (5 to 10 km) below the surface, where the fault is stuck instead of sliding. For example, the relatively fast-creeping southern Hayward Fault is only about 40 percent locked, on average, while the slow-creeping Rodgers Creek Fault is 89 percent locked, the study reports. When these locked areas build up a critical amount of strain, they break apart in an earthquake.

Lienkaemper and his co-author estimated a fault's future earthquake potential by combining creep measurements with mathematical fault models and other regional data, such as the time since the last earthquake.

The Hayward Fault has banked enough energy for a magnitude-6.8 earthquake, according to the study. The Rodgers Creek Fault could trigger a magnitude-7.1 earthquake, and the Green Valley Fault also has the potential to unleash a magnitude-7.1 shaker. The Northern Calaveras Fault is set for a magnitude-6.8 temblor.

Of all Bay Area faults, the Hayward Fault is most likely to spawn a damaging earthquake in the next 30 years, scientists think. Its 1868 earthquake was called the Big One until the great 1906 San Francisco quake came along. The Hayward Fault has ruptured about every 140 years for its previous five large earthquakes. The probability of a magnitude-6.7 earthquake on the Hayward Fault is 30 percent in the next 30 years.

Though 146 years have now passed since the last Hayward earthquake, that doesn't mean the fault is overdue for another quake, Lienkaemper said. "The average is 160 years, but the uncertainty is plus or minus 100 years, which is almost as big as the time [interval] itself." The 160-year average comes from an analysis of data collected from trenches dug across the fault that revealed evidence of earthquakes over thousands of years.

The Rodgers Creek and Green Valley Faults are also closing in on their average repeat times between earthquakes.

Email Becky Oskin  or follow her@beckyoskin. Follow us@livescience, Facebook  &Google+. Original article on  Live Science.

© 2012 LiveScience.com. All rights reserved.

Discuss:

Discussion comments

,

Most active discussions

  1. votes comments
  2. votes comments
  3. votes comments
  4. votes comments