Well I had an interesting day today. I took an excursion to various sites along the San Andreas fault with a group of Caltech alumni and seismologists, as well as explore the crawlspace underneath San Francisco City Hall where large rubber dampers support the entire building. Our guides were Dr. Carol Prentice of the USGS and Professor Tom Heaton from the Caltech Department of Engineering Seismology.
I learned quite a bit about earthquakes and earthquake safety I did not know about. We often hear such things like "We're 50 years overdue for the big quake," but such phrases are misleading. They are based on long-range averages but there is not any sort of clockwork or regularity between events. It is the norm for them to be either overdue or unexpectedly soon. Dr. Heaton noted that before 1906, there last two quakes on the fault occurred just 45 years apart in 1812 and 1857, whereas it's been a century since 1906. He also said that claims like "This building is constructed to withstand a magnitude 8 quake" are misleading as well....in fact, he says he's not sure what this is supposed to mean. An earthquake's magnitude only pertains to the amount of energy released, whereas how the energy is released makes all the difference to a building's structural integrity. The velocity of a building's movement over time is one key factor. If a magnitude 8 quake is done and over with in just a few seconds (with all that energy released together), the damage to a high rise building will be far less than the same quake stretched out over 30 or 40 seconds which would give the building enough time to attain a velocity (i.e. the sway of the top portion of the building away from the base) that can cause more significant damage...especially if the ground moves the base in the opposite direction that the top is going.
The really scary thing was Heaton's discussion of building safety. In short, our high rise buildings are far more dangerous in earthquakes than people realize. For one thing, up until the 1990s, welders produced much weaker welds (so-called "brittle welds") than are used today, and buildings that have these brittle welds fail much faster than expected in Heaton's computer models. High rises with so-called "perfect welds" perform much better but can still fail provided that the earthquake lasts long enough. Brittle welds should be replaced in buildings throughout California, but the expense and inconvenience involved in retrofitting will likely prevent it from happening across the board. He says that change usually does not happen until people die, sadly.
Ordinary houses, on the other hand, are among the strongest buildings. Wood frames are flexible and can still hold structure if damaged, and sheer-wall structures (when built according to our present-day building code, as opposed to older structures) while inflexible are very strong compared to the light loads involved in one or two-story buildings. A house may be torn apart and destroyed if it is on the fault line itself, but surprisingly even homes just a block away from the fault tend to be relatively undamaged. Heaton noticed that in all his years of experience, he has never seen a 7 11 convenience store suffer damage in an earthquake. Then, as we drove around in South San Francisco, we saw a 7 11 that was built almost directly on a fault line. Seeing that, he laughed and said that he might need to revise his generalization about 7 11s.
In the original Superman, Lex Luthor managed to start an earthquake by exploding a nuclear bomb along the San Andreas fault. Heaton actually noted that nuclear blasts can induce seismic shifts...he said that genuine earthquakes have occurred in New Mexico on the tail of nuclear tests. However, it is useless to think of using this or any other device (such as pumping massive amounts of WD 40 to induce slippage of the fault) to relieve pressure on the fault in order to avoid the catastrophic Big One. The Richter Scale is logorithmic, so in order to prevent a 9.0 Indian Ocean-type quake, we would need to have about 30,000 5.0 quakes to equal the same amount of energy. For some reason, having a daily 5.0 earthquake for 82 years does not seem to be a feasible proposition.
Dr. Prentice took us to the San Francisco watershed near San Bruno, which otherwise is inaccessible to visitors, and we hiked (through much poison oak) to a spot where the San Andreas fault had a surface rupture in the 1906 earthquake. Here is a photo of this spot from 1906, looking downhill:
Now before the earthquake, the fence and the row of trees on the left formed a straight line. The surface rupture can be seen clearly in the field, and the far end of the land has moved several meters north, bringing that portion of the fence and trees to the right. The trees and fence still stand today; in fact, this is the last genuine fenceline that still exists today...the ones on display elsewhere in the Bay Area are "reconstructions" (i.e. fakes).
Here are some pictures that I took of the fence line as it looks today:
Here I am looking up the hill in the opposite direction shown in the 1906 photo. You can see that the fence line in the foreground is further to the left than the portion that is further back, which is more to the right. As for the surface rupture that occurred 100 years ago, it is no longer visible on the surface as would be expected. Here is Dr. Prentice standing directly on the fault:
You might notice a slight depression in that location, with the land slightly higher behind her. That is the only obvious sign of the fault in that location, and you'd have to know that there is a fault there to notice it; otherwise, it just looks like any depression and rise. But just a few hundred feet away, there is further evidence of the fault: there is a creek that flows down to the fault and then abruptly changes direction to flow alongside the fault line, to again change direction as it crosses the fault. The path of the creek has been altered by the lateral displacement caused by movement of the Pacific plate along the fault.
Heaton and Prentice also took us to South San Francisco to see what happens when the fault line runs underneath a town. Redevelopment of the land has effectively erased all obvious traces of the fault which were evident after the 1906 earthquake. Here is a photo from 1906 of a locale in South San Francisco, where a displacement of several meters is evident from the broken fence line:
The nearer portion of the fence has moved north (i.e. leftward), but other fault lines are also visible further back. This fence stood at this site (which remained undeveloped) all the way until the late 1950s. But then it was torn down to put a power line through and then the land was bulldozed, reshaped, streets put in, and then houses built. Here is how the same site looked in 1962:
The power line is roughly in the same spot as the old fence and homes are being built right on top of the old fault line. The red arrow points to a house that still stands today, a little yellow home. Here is my picture of the same view, taken just earlier today:
The power line is still there on the right, and the two homes being built in 1962 are right there in the middle right....sitting squat on the San Andres Fault. As Professor Heaton (below) explains, it is today illegal to build on known fault lines. In fact, there is a park just out of view on the left that corresponds to the location of the fault and if you follow it further it continues to the parking lot of the 7 11. But these older homes, built before the law, are in use like any other home....despite the fact that they will surely be destroyed whenever the fault moves again. And as you can see in the 1906 photo, there are several more parallel fault lines further in the distance that similarly are built up.
Our last stop on the tour was the best. We got to go to San Francisco City Hall and see the shock absorbers that were designed to protect the building from the next major earthquake. The base isolation system is mentioned in the Wikipedia article on the building:
http://en.wikipedia.org/wiki/San_Francisco_City_Hall
The rubber-and-steel isolators effectively cut off the building from the ground movement, keeping it fairly still while the ground slides underneath it during an earthquake. On the surface all around there are steel slides under which the ground would move during a quake:
However, Heaton was somewhat reserved on how successful the system would be in an actual earthquake. He says that city officials and civil engineers are very confident of the system, declaring the building "safe" in an 8.0 or whatever quake, but Heaton pointed out that there is only so much distance the system can allow the ground to move, and if the sway exceeds this distance, the effect could be much like a ship driving itself into a wharf.
The neatest thing was being able to go UNDERNEATH the whole building to see the support system. Mind you, the whole City Hall is suspended on these isolators like springs, and there is a very narrow crawlspace below the basement that you have to climb down in a ladder through a hole in the basement floor. After signing wavers (having a building fall on you, for instance, would not be a good thing) and donning hard hats, we went three at a time into the tiny 2-3 feet high crawlspace. Here are some pictures I took of the shock absorbers:
When you're down there, you literally can see across the whole length of the building and see light on the other end. Very cool!
Another interesting thing that Professor Heaton said is that it is not known how well the Golden Gate Bridge would perform in "the Big One". City officials again assure us that it would be safe but Heaton noted that the bridge runs north-south and the fault moves north-south, so a lateral slip-strike movement could in theory swing the suspended deck into the towers...hitting them like a battering ram. But there is no way to know exactly what would happen until a real quake happens because there are many different variables that are unknown.
BTW, in case anyone wondered, I asked him what he thought of the idea of some that "earthquakes increased in number in the 20th century," and he said -- as I fully expected -- that the notion is totally wrong and is based on a misunderstanding of the representativeness of earthquake catalogues....records were fairly sparse until the 20th century.