May 20, 2014
Earthquake Science: Predicting The Next ‘Big One’ For San Francisco
Lawrence LeBlond for redOrbit.com - Your Universe Online
The Great 1906 San Francisco earthquake released as much accumulated stress as a cluster of closely timed temblors did over a 100-year period in the seventeenth and eighteenth centuries. Based on this historical data, new research published in the Bulletin of the Seismological Society of America is looking at two possible scenarios for the San Francisco Bay Area’s next “Big One.”
The San Francisco Bay Region (SFBR) sits on a boundary between two major tectonic plates: The Pacific and North American plates. When earthquakes occur, the energy is released along principal fault lines that exist in the region – these include the San Andreas, San Gregorio, Calaveras, Hayward-Rodgers Creek, Greenville, and Concord-Green Valley faults.
"The  earthquake had the beneficial effect of releasing the plate boundary stress and relaxing the crust, ushering in a period of low level earthquake activity," Schwartz said in a statement.
The quake cycle reflects the accumulation of stress, its release as slip on a fault or a set of faults, and its re-accumulation and re-release. The historical data shows that the SFBR has not experienced a full earthquake cycle since it has been occupied by reporters of earthquake activity, either through written documentation or instrumentation. Records of earthquakes and earthquake damage have been kept by the Mission Dolores and Presidio in San Francisco since 1776, the farthest back for which reliable records exist.
"We are looking back at the past to get a more reasonable view of what's going to happen decades down the road," said Schwartz. "The only way to get a long history is to do these paleoseismic studies, which can help construct the rupture histories of the faults and the region. We are trying to see what went on and understand the uncertainties for the Bay Area."
Schwartz and his colleagues excavated trenches along faults, observing past surface ruptures from the more recent earthquakes in the area. They used radiocarbon dating of detrital charcoal and measured the presence of non-native pollen to establish the dates of paleoearthquakes, expanding the span of data of large quakes back to 1600.
The trenching studies suggest to the team that between 1690 and 1776, a series of earthquakes ranging in magnitude from 6.6 to 7.8 occurred on the “Hayward fault (north and south segments), San Andreas fault (North Coast and San Juan Bautista segments), northern Calaveras fault, Rodgers Creek fault, and San Gregorio fault.” The team was not able to pull paleoearthquake data for the Greenville fault or the northern segment of the Concord-Green Valley fault for this time period.
"What the cluster of earthquakes did in our calculations was to release an amount of energy somewhat comparable to the amount released in the crust by the 1906 quake," said Schwartz.
As stress builds up in the region, the energy will release either of two ways: either a massive earthquake like the 1906 Great One, or a cluster of large earthquakes, as was seen in the century before earthquake records were first kept in 1776.
"Everybody is still thinking about a repeat of the 1906 quake," said Schwartz. "It's one thing to have a 1906-like earthquake where seismic activity is shut off, and we slide through the next 110 years in relative quiet. But what happens if every five years we get a magnitude 6.8 or 7.2? That's not outside the realm of possibility."
Roland Burgmann, an earth scientist at UC Berkeley agrees with study.
The various faults in this region “talk” to each other. “The communicating family of faults sometimes tend to rupture together as a group or shut each other off,” Burgmann told Mercury News’ Becky Bach.
Schwartz said the 1906 SFBR quake was likely a fluke, the perfect alignment of conditions that allowed 300 miles of the San Andreas Fault to release its pent-up pressure. That “Big One” kept the area relatively quiet for the better part of the next century, he added.
"Eventually, there should be more clusters," Burgmann told Bach.
Although Schwartz and colleagues predict the next “Big One” will be a series of tremors, rather than a single massive earthquake, they acknowledge that they could also be wrong.
The Bay Region has a 63 percent chance of one or more large temblors before 2036, according to estimates released by the Working Group on California Earthquake Probabilities in 2008.
Schwartz told Mercury News that he wants to peer farther back into the past to gain a better understanding of the SFBR’s seismic history in hopes to be better able to predict the region’s future.
"This study represents a heroic amount of work," Greg Beroza, a professor of earth sciences at Stanford University in California, told NBC News. "The basic problem in forecasting earthquakes is that we have a very limited record and an incomplete understanding of how earthquakes work."
He noted that Schwartz and colleagues’ paleoseismology work required a huge physical effort, yet it only works for large, surface-rupturing earthquakes. In essence, they are recording only the “conservative lower bound.”
"The data is limited — dates are uncertain as are magnitudes — but despite these limitations, they have shown that the Bay Area was very seismically active during this period,” he told NBC’s Joel Shurkin.