CALIFORNIANS bracing themselves for the next 鈥渂ig one鈥� may not have long to wait. The most comprehensive set of studies ever carried out on the San Andreas fault all drive home the same grim finding 鈥� large areas are overdue for a major quake.
The 1200-kilometre fault runs under some of California鈥檚 most densely populated areas, and a big quake of magnitude 7 or above could easily disrupt the lives of 5 to 10 million people. Though most buildings in the region are designed to reduce the death toll in the event of a disaster, the economic consequences for areas such as Los Angeles or Silicon Valley are likely to be catastrophic, with ripple effects worldwide. Californians already live with the knowledge that a big quake is inevitable, but the new reports, the result of decades of work by 14 different teams, confirm that in both northern and southern areas of the state, the time since the last major quake is longer than the average time between quakes.
Previous risk estimates were based largely on written records of earthquakes, but in California these go back only two centuries. Now, geologists have borrowed techniques from archaeology to push back those time lines. They have excavated 10 sites along the San Andreas fault where quakes have broken the surface, and used radiocarbon dating to analyse sediment layers surrounding disruptions caused by past tremors. The results appear in a special issue of the Bulletin of the Seismological Society of America.
Advertisement
Thomas Fumal of the US Geological Survey (USGS) in Menlo Park, California, and colleagues studied the southern part of the fault, near Ontario and San Bernadino. By matching disrupted layers unearthed in 38 trenches, they dated 14 large earthquakes from the past 1500 years 鈥� the longest quake sequence ever made in the US.
The last major quake in the area was 145 years ago, in 1857. But over the past 1500 years, the average gap between big quakes has been just 105 years. Fumal suspects that a long stretch of the southern San Andreas will break within the next few decades, unleashing a quake of magnitude 7.6 to 7.8 鈥� large enough to make itself felt all over southern California. Independent experts estimate it would cost the area $50 billion.
The team also found that a wider area may be affected than previously thought. The ancient quakes disrupted widely separated fault segments, undermining the assumption that just one segment breaks at a time. 鈥淭he data suggest larger earthquakes,鈥� Fumal says. 鈥淭ypically the fault ruptures as it did in 1857 鈥� for 320 kilometres.鈥�
Geophysicist Jim Lienkaemper, also at the USGS in Menlo Park, found a similar scenario in northern California. His team studied the Hayward fault, which lies under the densely populated east side of San Francisco Bay, and identified four major quakes since 1470. The average gap between them is about 130 years, less than expected. The Hayward fault last ruptured 134 years ago in 1868.
Data from the studies will help revise the regional hazard maps that planners and engineers rely upon, and may also affect risk estimates in Turkey and New Zealand, which are crossed by similar faults. However, some seismologists question whether it makes sense to try to predict when the next quake is likely to occur. 鈥淚t鈥檚 quite an inference to say we鈥檙e overdue,鈥� says Lucy Jones, head of the USGS Pasadena office. 鈥淲e don鈥檛 have any data that tells us earthquakes 鈥榗are鈥� how long it鈥檚 been since the last quake.鈥�
Lienkaemper agrees that quakes do not occur with perfect regularity. 鈥淏ut the real message,鈥� he says, 鈥渋s that there鈥檚 a whole group of faults that are overdue. We should be preparing.鈥�
