Richard Marsden graduated with honors from Churchill College, University of Cambridge, with a degree in Natural Sciences (Geological). He also holds a Masters Degree from the University of Durham. It was as part of this that he worked with Drs. Mary Lou Zoback and Bruce Julian at the U.S. Geological Survey, Menlo Park, California, on the Daly City Earthquake of 1957. A paper about the earthquake was presented to the Fall 1995 meeting of the American Geophysical Union. The abstract can be found on his homepage.

As a keen amateur astronomer, Richard served on the Committee of the Cambridge University Astronomy Society in its 50th and 51st years.

Richard is also a major author of the CADIE project - an educational computer tool for diabetics. He currently works and lives in Houston, Texas.

On 22 March 1957, the San Francisco Peninsula was rocked by an earthquake centered in the Daly City area that caused damage in Westlake District of that city, and some damage in San Francisco. Although only magnitude 5.3, this was the largest earthquake on the Peninsula since 1906.

Recent work which has attempted to find the epicenter of the 1906 San Francisco earthquake has suggested an epicenter on or near the Peninsula.

Originally, this epicenter was thought to be at Point Reyes, but work using distant seismograms suggests a more southerly epicenter.

Because it now appears that the Daly City earthquake occurred in the region where the 1906 earthquake was centered, the 1957 event suddenly becomes more important for scientists, as well as the people who live and work in the Bay Area.

Unfortunately, the seismometers then in place in and around the San Francisco Bay were not as numerous as they are now, so the earthquake mechanism data derived from 1957 seismic recordings were not conclusive.

By analysis of seismograms after an earthquake, it is possible to discover the orientation of the plane. Usually, it is assumed the earthquake occurred at a point - the focus. The epicenter is the point lying on the Earth’s surface, directly above the focus. Because the earthquake mechanism is just for this point, the mechanism is often called a “focal mechanism.” This is extremely useful, because scientists can then characterize particular earthquakes. This, in turn, allows us to work out why earthquakes happen.

There are three main kinds of earthquake mechanism:

1.) Strike-slip (Transform or “Sliding”) Faults
In these, the fault plane is vertical or near vertical, and the motion of one block is in a horizontal direction parallel to the plane. The most famous example is that of the main San Andreas Fault. In the 1906 Earthquake, the west (Pacific) side moved north relative to east (California) side. There is little or no vertical movement on the fault, although the surface may show local slumping and slippage.

These faults typically occur when a plate is moving sideways relative to an adjoining one (eg. Pacific and North American Plates). They also occur near mid-ocean ridges (eg. central Atlantic), and combined with other fault zones (Kobe 1995 occurred on a transform fault).

2.) Compressional (“Thrust”) Faults
These occur where forces are pushing two blocks of rock together. The fault plane will be dipping, with the upper (“Hanging Wall”) block overriding the lower (“Foot Wall”) block.

These faults typically occur in mountain ranges where the thrusting of blocks causes mountain uplift (eg. Himalayas and Alps), or in subduction zones (eg. Japan).

Due to the compression, the crust tends to thicken, hence it is no coincidence that most areas of compressional faulting are of relatively high terrain.

3.) Extensional, or “Normal” Faults
These occur where rocks are being stretched. Again, there is a dipping fault, but the upper (“Hanging Wall”) block moves down. These are often likened to dominos leaning against one another. This effect can be seen in Nevada, where the faults form the steep sides of the mountain ridges.

Due to the stretching, the crust tends to thin, so areas experiencing extensional faulting tend to be low-lying. About 100 to 200 million years ago, this occurred in the North Sea. This provided an ideal environment for oil formation, and the area is still below sea level. Extensional faults are also very common along mid-ocean ridges.

Initial work immediately after the 1957 earthquake showed that this was unlikely. Unfortunately, the earthquake was limited in size, so records were limited.

During the summer on 1995, I worked with Dr. Mary Lou Zoback and Dr. Bruce Julian from the U.S. Geological Survey at Menlo Park, with help and using methods developed by Dr. Doug Dreger at U.C. Berkeley. We attempted to get all the records still extant. These data were combined with modern seismic-wave modelling methods, applied to the better seismograms, to so reduce the number of possible earthquake mechanisms.

We decided that the most likely mechanism was that of an extensional event.

Isn’t this a bit strange?

At first sight it might be, but there are a number of pieces of evidence that tell us that we should expect some extensional faulting near Daly City:

1.) In recent years, a dense network of seismometers has operated in the San Francisco Bay region. This network picks up hundreds of small earthquakes — the vast majority too small to feel. Along the San Andreas fault, two clusters are of note. In the Santa Cruz mountains, there is a cluster of compressional events related to the 1989 Loma Prieta earthquake. Another, smaller cluster, lies in the Daly City. These are predominately extensional faults.

2.) The San Andreas fault is not dead straight, but has a number of small kinks. Curving of strike-slip faults would be expected to lead to compression or extension faulting. In the Santa Cruz mountains, the fault curves slightly to the left, whilst along the Golden Gate section, it curves slightly to the right. The Pacific plate is moving northwards, so it “hits” the American plate in the Santa Cruz mountains, but moves away from it in the Golden Gate region. This is why compression faults are seen in the Santa Cruz mountains, whilst extensional faults are seen in the Daly City region. In fact, the Loma Prieta earthquake was caused by a combination of compressional and strike- slip motions.

3.) This also explains the topography. Compressional faulting occurs in the Santa Cruz mountains, hence the crust thickens, and mountains form. In the Daly City region, faulting is extensional, so the crust thins, hence the fault drops below sea level.

The original is from a Benioff seismometer (T0=1sec, Tg=90sec) recording made at the PAS (Pasadena) station 540 km [336 miles] from the epicenter. This may sound like a long-period instrument, but it is actually one of the shorter recordings that could be found. The original is a paper (analog) record, later scanned at the University of California Berkeley, and digitized at U.S.G.S., Menlo Park.

This recording is unfiltered, but speeded up 134 times—relative to real-time—to make it audible. The recording, therefore, does not sound very realistic with small PC speakers. It does sound surprisingly authentic when played through a medium-to-low-end high-fidelity amplifier. For best effect, set the bass control to “full.” The results should allow you to hear the recorded 1957 earthquake vibrations from the eaves of a four-story building, to the basement.

You may also experience the 1957 earthquake at the shake-table located at the California Academy of Sciences’ Museum of Science and Natural History in Golden Gate Park.