Parkfield: Biggest Little Town in Seismology
Parkfield California is not a town that you’re going to stumble upon while headed somewhere else. With a population of just 18 (circa 2018), it doesn’t even appear on many maps. It was not in the database of my Garmin mobil navigator.
But if you’re a science nerd who’s into earthquakes and seismology, chances are pretty good you’ve heard of it. I had read about Parkfield for years, and was eagerly anticipating a chance to visit. In his excellent book, A Crack in the Edge of the World, Simon Winchester spends a good deal of time discussing Parkfield and its place in earthquake history.
The main reason I wanted to visit was to see this bridge. Located just south of town, it carries the Parkfield-Coalinga Road across the Little Cholame Creek. The creek also marks the surface trace of the San Andreas Fault. It’s generally a dry creek-bed but there was a little trickle of water on the day I visited (September 7, 2018). But first, the backstory.
In the latter part of the 20th century, the science of earthquakes and seismology was ramping up, and California was a hotbed of activity due to the many faults (and quakes) that occur there. Scientists had noted that moderate-level earthquakes (magnitude ~6) had occurred in or near Parkfield in 1857, 1881, 1901, 1922, 1934 and 1966. With an average period of about 22 years, they figured the area should see another quake in the late 1980s or early 90s. While the history of predicting earthquakes was a mashup of bad science and outright quackery, here was a chance to apply some real science to a geographic area that seemed to be experiencing real regularity in the spacing of its earthquakes and the behavior of the fault.
The San Andreas fault behaves differently along various segments of its 750 mile trace through California. Along the northern and southern regions (e.g, San Francisco and LA), the fault “locks up” and accumulates stress and strain that is eventually released in the form of large earthquakes. The 1906 San Francisco quake was an example of this. The center section — which includes Parkfield — is different though. Here, the fault moves constantly in what seismologists call “aseismic creep,” where the fault slips more or less continuously without causing large earthquakes. The smaller quakes, with the relatively frequent turnabout, provided scientists with the opportunity to turn Parkfield into an experimental earthquake observatory of unprecedented sophistication.
Starting in the late 80s, scientists from the USGS and University of California have installed millions of dollars worth of sensing equipment in — and under — Parkfield. This was done, and the waiting commenced.
The early 90s came — and went. No earthquake was forthcoming. Not a problem though, because the 22 year period was based on a history of six known quakes and was expected to come with significant error bars. Earthquake prediction, as noted, is not an exact science, but just a probability based system of observation.
The much anticipated Parkfield quake finally came on September 28, 2004, when a magnitude 6.0 quake struck the fault near the town, and under the watchful eye of all the creep-meters, lasers and ground-motion sensors. You can read more about the Parkfield observation program here, here and here. It’s a great story and a good example of science done right. Though the experiment’s main hypothesis was not confirmed (there were apparently no ground electrical anomalies that could be regarded as precursors of the earthquake), it was nevertheless the most monitored and surveilled earthquake ever studied, up to that time.
Back to my visit though. It’s really all about that bridge. There are only a handful of places in the world where you can see evidence of fault deformation clearly at the surface. In most places along its trace, the San Andreas fault is buried under deep sediments, or masked by tree cover and desert, or urban development. In a few places (Wallace Creek at Carizzo, or the Palmdale road cut) you can see dramatic and indisputable visual evidence, but for most of its nearly 800 mile run through California, you’d never know it’s there unless you were a geologist.
Which is why I came to Parkfield on this lovely September day. It’s just a bridge over a creek but it’s enough to make this tiny town a tourist attraction for many. That right-hand jog that the rail takes as you cross the bridge from east to west happens because the entire Pacific plate is slowly creeping to the north. (Thus the San Andreas is known as a “right lateral” fault. No matter which side you stand on, the opposite side seems to be moving to the right)
Since the bridge spans the fault itself, it presents both short- and long-term engineering challenges. The west side of the creek (the Pacific plate) is creeping to the north at a rate of about an inch a year. This creep manifests itself in a skewing of the bridge itself. As seen in the photo above, the railing is bent to the right as one sights down the length of the bridge. In this view, I’m standing on the east side of the creek, on the North American Plate. The far side of the creek (the Pacific plate) is moving to the right, and carrying the far end of the bridge with it. All bridges need periodic maintenance and repair, but it’s not hard to imagine that this particular bridge presents some additional challenges to the California DOT.
I contacted the CalDOT and an engineer was happy to reply to my query about the Parkfield bridge. They were very familiar with it but, surprisingly to me, indicated that it needed little in the way of additional monitoring. Apparently, in California, seismic movement of infrastructure is just part of the job. The engineer said that all bridges normally get inspected every two years; bridges with active fault deformation may get more frequent inspections, but the Parkfield bridge is currently in good shape and performing as expected, so it merits no special attention right now:
While the displacement of the fault and the effects to the bridge are clearly visible, the substructure and superstructure appear to performing as intended by accommodating the fault creep with no significant structural distress.
California DOT knows how to do bridges!
This is the expansion joint located near the mid-point of the bridge. You can see the extra spacer installed on the south side of the bridge to allow for the torquing to the north.
The above view is looking back toward the east along the other bridge rail. For perspective, the bridge spans about 200 feet over the Cholame creek, so the rail you’re looking at is 200 feet long. In this image, the foreshortening due to perspective exaggerates the skew a little, but it really is quite striking, knowing that engineers instinctively like their lines to be straight. Look also at the abutments resting on the bedrock under the creek bed. One presumes that those were originally surveyed to be straight also, but mother nature — and the San Andreas fault — had other plans.
I went to some trouble to visit Parkfield. As I said out the outset, it’s not on the way to anywhere else. You have to really want to visit Parkfield, and I really wanted to. I wanted to see this bridge with my own eyes. I wanted to walk across the plate boundary. I wanted to stand with my left foot on the North American plate and my right foot on the Pacific plate (you can’t really do this and since there was water in the creek, I didn’t try). I just wanted to be there.
And so I went. And it was soooooo satisfying. It doesn’t take much to please me. I got to Parkfield at around 9:30 in the morning. There is a cafe to cater to the town tourism (related to ranching and earthquakes) but it didn’t open until 11:30. I sat on the porch in the increasing heat of the California day and read and contemplated the next stops on my San Andreas field trip. And when the cafe opened, I had lunch.
It was a great day!
For further reading on Parkfield, CA’s earthquake research:
USGS site on the Parkfield experiment
Another USGS page on research at Parkfield
Journal article on the findings of the 2004 Parkfield quake
Brief popular article discussing implications of the 2004 quake findings
Nature article discussion the 2004 quake (paywall)
