In the News (Again)

Today’s Chronicle carried a front page article on earthquakes in California that stated we are 99% certain to have big (6.7 or higher) earthquake in the next 30 years.

Forgive me if I yawn here, but that information is not only obvious from past occurrences, but does not help anybody, wherever they live. It’s like saying that there is a 99% certainty that dawn will break tomorrow morning.

In order to arrive at this prediction, the “scientists in the group used complex analytical tools…and powerful new computer programs”. “Computer power played a large part in developing the new probability models” said geophysicist Tom Parsons of the USGS in Menlo Park, a member of the group that produced the report.

The article went on to say that “the probability that a magnitude 6.7 quake will hit on any one of the Bay Area faults is 63 percent, only a tiny bit higher than the 62 percent estimated by a similar group in 2003. But the probability for that kind of severely damaging quake on the Hayward-Rodgers Creek fault was increased in the new forecast from 27 to 31 percent”.

It may be that the article was poorly written and researched, but it left me thinking “where’s the beef?”

The fact of the matter is that the Hayward Fault last split open with a surface rupture of 6 feet horizontally exactly 140 years ago on October 21, 1868, with an estimated magnitude of 7.0. Archaeological digs have found the evidence that the average interval between the last five earthquakes on the Hayward Fault is 140 years. But that was on the southern part of the fault, in the Hayward-San Leandro area.

I don’t need complex computer programs to tell me that the probability of another 7.0 quake on that part of the fault within the next few years is pretty close to 100%. But how can we get a handle on the northern part which runs through north Oakland, Berkeley, and the cities to the north?

Doris Sloan, an Adjunct Professor in the Department of Earth and Planetary Science at UC Berkeley says in her 2006 book Geology of the San Francisco Bay Region that “evidence from the trenches dug across the fault at the Mira Vista Golf Course east of Richmond indicates that the last rupture of that segment was between 1640 and 1776. Because the northern Hayward Fault has not moved in a long time, the probability for an earthquake may be higher for this segment than elsewhere”.

You can read more details of the Mira Vista Trenching Study in the paper Timing of Paleoearthquakes on the Northern Hayward Fault. If you try to sort through the details and nail down some dates, it would seem that the last earthquake on this part of the fault was somewhere between 1640 and 1776 as Doris Sloan points out, and the average recurrence time is between 270 and 710 years. Assume the worst (the shortest interval), and that puts the next earthquake between 1910 (which clearly didn’t happen) and 2046. We are already more than two thirds of the way through that interval.

This is pretty flaky logic, because that study has pretty flaky data. However let’s revisit that Chronicle article and repeat a quote from earthquake geologist David Schwartz of the USGS. “The further you are in time from the last quake on a fault, the higher the probability is for the next one”.

Visit my recently developed Fault Walk (Walk 1 at the top of this page) to see photographs, or walk the fault itself.