Some anti-nuclear sites are using an earthquake map from the LA Times on April 28, 2010 that show the Newport-Inglewood fault in red, indicating an earthquake potential of 7.5 – 8.0. I didn’t know whether to flee first, or take the time to look into it first. Remember, with great power of earthquakes comes great responsibility to be accurate. I rembered that earthquakes come with certain probabilities per year, or certain frequency, and decided to find out what that was associated with these dire magnitudes. The Uniform California Earthquake Rupture Forecast Version 2 (UCERF 2) published in 2009 in the Bulletin of the Seismological Society of America is the source of this data. It classifies 6 faults as Type A including the San Andreas fault, and all others as Type B, including the Newport-Inglewood fault. It treats all California Type B as having the same probability per year curve versus magnitude, with two differing models. If you are trying to forecast the probability of any serious earthquake occuring, this is a simplifying model of treating all the same. However, it throws out a lot of unique information about any individual fault, such as creep rate and historical frequency and magnitude of rupture, which is very important for predicting that given fault. Even within the uniform approach however, we will find that labeling the Newport-Inglewood as 8.0 and even 7.5 is really distorting the truth in its probability of rupture.
Lets start with the 8.0 quake probability. On the Cumulative part in Table 9, p. 27 of the download, the rate is one per 40,000 years (I round off). This means one out of 500 of the probability of such an earthquake during a 80 year lifetime. Is it informative or seriously misleading to label the Newport-Inglewood fault with an 8.0 on a map? I blame this on the LA Times, not on the groups that just quote it.
There are graphs on p. 30 of the download of UCERF 2 (labeled p. 2082), and we will use the one for Southern California.
Lets now examine the still hazardous 7.5 quake probabilities and labeling. The combined result is 1.3×10^-3, or one in 800 years. Larger, but at most one in 20 in a 40 year plant lifetime, or one in ten for an 80 year person’s lifetime.
For a magnitude 7.25 quake from the graph, we have one in 140 years. For a magnitude 7.0, the estimate is one in 40 years. The latter is now in the timescale of the San Onofre plant license, and the magnitude for which the San Onofre plant is strengthened for. However, they have a much smaller estimate of the probability, since they claim that the nearby faults have not ruptured in 120,000 years.
However, we mentioned at the start that one should actually look at each fault separately. I don’t have trenching results, and have to stress again that I am not a geologist. The years of the largest quakes of the Newport-Inglewood fault are 1769, 1812, 1855, and 1933. The intervals are 43, 43, 78, and now 79 years without an extreme quake. The magnitude of the 1933 Long Beach quake was 6.4 or 6.3, and the intensity of shaking that determines damage was the very serious VIII. Brick buildings fell as well as wooden ones. One hundred and fifteen people were killed in that quake. Seventy schools in Long Beach and Compton were destroyed, and 120 were heavily damaged. Fortunately, nobody was in school at the time of the quake. In a month, the Field Act was passed requiring earthquake restraints in school construction. People lived in tents because of after shocks. On a personal note, my Father lived through that quake in Long Beach. I was born in Long Beach in 1941. Last week at lunch with my wife’s friends, three also remembered that quake. The largest quake on the fault in the intervening years has been a 4.7.
While the general model of Type B faults estimates a 7.0 quake every 40 years, for the actual Newport-Inglewood fault, for the last 157 years, there was only one large quake, and it was a 6.4, not a 7.0. This is far below the general model in probability and magnitude. By the general model we should have had a 7.25 in just 140 years. In view of this, I think even the 7.5 label on the map for Newport-Inglewood is not justified. Wikipedia agrees, listing the strength of the fault as a 6.0 to 7.4.
We will just have to wait for the earthquake hazard survey for the specific San Onofre site, and not use the general, one size fits all model.
The creep rate for the Newport-Inglewood is quoted as 0.6 mm per year. The slip rate of the nearest segment of the San Andreas fault, the San Jacinto fault, is 13-18 mm per year. This is about 20 to 30 times faster than the Newport-Inglewood fault. If we think of creeping for the last 80 years on the Newport-Inglewood fault without a major quake, we get a 48 mm or 4.8 cm or 1.9 inches of stress built up.
There is some confusion about the relation of a 7.4 quake, like the recent Mexican one, to a 7.0. For every 0.2 increase on the magnitude scale, the energy released in the quake goes up by a by a factor of two, so the difference of a 7.4 from a 7.0 is an increase by a factor of four in energy.
However, in terms of shaking, the increase in strength is 10^(0.4) = 2.51, which, of course, is still quite significant. The Mexican quake was on a subduction fault of the Cocos Plate under the North American plate.