The tiny town of Parkfield, California is the closest thing in the United States to a living seismic laboratory, a place where the United States Geological Survey and other organizations test out the best theories of earthquake origin, prediction, preparation, and response. On the surface Parkfield is a quiet ranching town with 18 permanent residents at the northern end of the Cholame Valley, but it’s also a town that sells itself as the Earthquake Capitol of the World.
The truth is there’s not much to see. But beneath the surface Parkfield is constantly rumbling with earthquakes, little ones all the time and bigger ones hitting more than most people would care to experience. Though it may not have the greatest quantity of earthquakes, it has perhaps had the greatest number of quality earthquakes—somewhat predictable big quakes that can be scientifically studied. Parkfield is probably the most mapped and measured spot along the notorious 800 mile long San Andreas Fault.
The San Andreas is a transform, or strike-slip fault defined by movement of the North American and Pacific plates as they grind against each other. The Pacific plate moves roughly northward while the North American plate drifts southward. Every now and then this slow tense grind lets loose with a dramatic slip and an earthquake occurs. The Grand Tejon quake of 1857, a truly massive 8+magnitude temblor, locked up the Parkfield section of the fault, creating a zone of nearly constant grinding seismicity. Beginning in the late 60’s and early 70’s the USGS set up shop in Parkfield, rigging the landscape with a vast net of seismometers, tilt meters, gas sniffers, cameras, and more recently GPS and satellite technology, all in an effort to “capture” a 6+ magnitude earthquake.
“Capture” suggests an effort akin to catching lightning in a bottle, an attempt at harnessing chaos; and the plan was to “capture” as in frame, freeze like a photograph, a specimen to be studied, dissected and mapped, as if they were neuroscientists charting the brain-waves and behavior of their most cherished and confounding patient, this eternally seizing epileptic planet we call home.
The plan, dubbed the Parkfield Experiment, was to wait for a big quake to hit the region–something that seemed fairly easy to predict based on the abundance of seismic activity in the valley–and to use their monitoring equipment to record every aspect of its behavior, thereby providing the USGS with a comprehensive map of the quake’s personality. They wanted to see all the foreshocks, fissures, P-waves, S-waves, amplitudes and aftershocks. The plan, to some people, was ambitious, and perhaps a bit crazy. The plan was to capture a quake that would allow an unprecedented picture of an earthquake and provide the kind of data that could refine the admittedly rudimentary science of earthquake prediction. The plan was not to cause an earthquake. The USGS would tell you that this is nearly impossible. They would tell you that humans are just too insignificant to affect the seismicity of our planet.
In the summer of 2010 the USGS office in Parkfield was housed in a brown singlewide mobile home parked next to the Cal-Fire station, not far from the Parkfield Café and the Parkfield Inn. I drove down one morning from my home in seismically silent, Fresno for a meeting with Andy Snyder, the man largely responsible for supervising the Parkfield experiment for the last six years. By the time Snyder arrived in in the Cholame Valley in 2004 for an ambitious new phase of the Parkfield Experiment, the USGS had been waiting almost forty years to capture their ultimate target earthquake, a big 6+-magnitude rumbler. It had been a long wait, so long in fact that priorities had begun to shift, methods had begun to change. Andy Snyder was the man charged with supervising these new priorities and new methods. No longer satisfied just sitting around waiting for a big quake, the USGS had decided to go after them where they lived.
When I met Snyder, he drove a black Mazda Miata with a vanity plate that read, “EQHAZRD,” which he parked beneath the shade of a small carport just steps away from the narrow brown modular trailer where he spent much of an average day staring at a computer, monitoring the vast array of sensing equipment spread around the Cholame Valley. Snyder watched it all, but paid special attention to data coming in from his pet, his baby, a truly monumental undertaking that defined the new priorities in Parkfield, something called the San Andreas Fault Observatory at Depth, or more colloquially, SAFOD.
Snyder was neither tall nor short and with his sandy blond hair, ruddy complexion and blue jeans looked like he’d be comfortable parked at a beachfront bar or hitting golf balls on Sundays. Snyder didn’t live in Parkfield, couldn’t live here, he told me. Instead he commuted from San Luis Obispo, and liked it that way because the town had things like restaurants, grocery stores, farmer’s markets, wine-bars, movie theatres and art galleries.
“It has culture,” he said with an audible wink.
Snyder took over in Parkfield after working in the oil and gas industry in what he called, “exploration.” Snyder wasn’t a geophysicist. He was a driller, oil mostly. He was an engineer, not a scientist. That’s why the USGS brought him here to supervise SAFOD, a massive drilling effort funded at least in part by the National Science Foundation. The ambitious goal of the project was to drill a three-foot diameter shaft directly into the San Andreas fault itself, and to not only monitor seismic activity as close to the source as possible but also to bring back core rock samples from the fault for scientific study. The plan was to go after earthquakes in their nest, to study them at the source.
If you’re like me, when you hear about something called the San Andreas Fault Observatory at Depth, then you dwell on that word “observatory,” and imagine yourself in a sealed-off glass enclosure descending into the earth on some kind of funicular track with a tour guide in a uniform. You imagine yourself staring into a chasm or cross section of rock that reveals the San Andreas Fault in action. You hear “observatory” and expect a tour guide’s measured delivery of facts mixed in with the pull of human-interest stories, and you begin to believe that you’re in for some kind of hands-on experience where you engage with the shifting landscapes of geological time in meaningful and significant ways. You might even expect to buy a t-shirt in the gift shop.
This is not SAFOD.
Snyder had tried to warn me. In a tersely worded email he said Parkfield wasn’t what he’d call a “science travel” destination. 
When I asked him that day if I could visit SAFOD he looked at me like I was stupid, paused and said, “It’s just a shack and a slag pit.”
“Um, OK,” I said, nervously rubbing my hands on my knees. My dreams of seeing the San Andreas died with a whimper.
Snyder swiveled back and forth in his desk chair and then kept talking. “Right now we’re dealing with the problem of what to do with all the stuff we took out.”
I admit I was disappointed there was nothing to see, nothing to visit. SAFOD, I learned, was a monumental project with no monument, an observatory hidden away on an isolated section of the San Andreas on private property, totally inaccessible to the public. The only thing you could really observe from SAFOD was sensor readout on Andy Snyder’s computer.
I’d come to Parkfield hoping to see the San Andreas, perhaps to feel something big like an earthquake. But I’d expected too much.
“The fault itself is only a couple of inches wide in some places,” Snyder told me, holding up his fingers to show me the gap. “It’s surrounded by looser material, but the crack itself isn’t much at all.”
“Sure, but ‘s 10 miles deep and 800 miles long.”
“Ten miles deep?”
That’s 52,800 feet, or nearly 25,000 feet higher than the tallest mountains on the planet and 20,000 feet higher than the altitude at which passenger jets fly. Snyder showed me giant educational poster boards that stretched nearly from one side of the trailer to the other and illustrated in bright colors how the SAFOD drill-rig punched it’s three-foot and descending diameter shaft three miles, or almost 16,000 feet down into the earth’s crust, bending at forty-five degrees right before plunging straight into the San Andreas Fault itself like a needle taking a biopsy.
I asked Snyder if the USGS worried about the danger, the potential seismic impact of drilling a hole that wide and deep into the most active fault zone in North America.
“Sure, we thought about that,” Snyder said, “but it’s the equivalent of sticking a push-pin into that wall.”
He pointed to the wall of his trailer, chuckled, and returned to his giant posters.
I laughed, too, but I was still staring at the wall and thinking: inches wide and miles deep. I was thinking about hundreds or thousands of pushpins shoved into a crack in the wall. I imagined the biggest push-pin in the world, a huge cartoonish red thing plunging into a fragile egg laced with cracks and fissures covering a boiling core, a shell that floats around on that center, moving on a geological clock, apparently independent and inconsiderate to the efforts of human survival. Or maybe not. Maybe human life and the life of our planet are more intimately connected than we truly understand. Maybe when we poke and prod, drill and blast into the earth, we unleash consequences we can’t predict.
Soon after drilling at SAFOD was coming to an end in 2004, just as they were settling into the next monitoring phase of the Parkfield Experiment, at 10:15 in the morning of September 28, a magnitude 6 earthquake rippled up from nearly 5 miles beneath the surface and rumbled through the valley, the biggest Parkfield earthquake in nearly 40 years. As the USGS watched and listened, the quake shook long enough (nearly 10 seconds) and hard enough to become the most “captured” quake in the history of the Parkfield Experiment.
It was kind of a big deal. A resounding success. But few people outside of Parkfield or the somewhat insular community of seismologists and earthquake obsessives knew much about it. Nobody was killed or injured. There was a little property damage. Life in the Cholame Valley proceeded mostly uninterrupted, just as it always does.
When I visited some six years later with geophysicists, Andy Michael and Diane Moore at the USGS Western Regional Headquarters in Menlo Park, California, I asked again if anyone living in the Cholame Valley in 2004 had raised concerns about the drilling at SAFOD and the 2004 Parkfield quake. I was basically asking if anyone blamed them for causing the quake they were happily studying.
We stood in a hallway staring at brightly colored posters on SAFOD. The USGS likes posters and maps, and I’d already spent a delirious few minutes collecting maps and other documents from their sales department before wandering around gawking at the poster and map-filled walls of the building. The halls were like a gallery of seismic art.
Moore, a woman who exuded the calm wisdom of a veteran high school science teacher, was the first to respond to my question. “What are there like 20 people living in Parkfield?” she asked. “We’d probably get more complaints from cows.”
She chuckled a bit and smiled at me.
“Yeah, right,” Michael interjected, “but think about the difference if SAFOD was in a major metropolitan area? Think of the impact on public perception.”
“Sure, like if we’d drilled into the Hayward fault?” Moore asked and shook her head. She suddenly looked like she was giving me bad news about my grade in her class.
We all stared at the poster. And I understood what they were saying. If SAFOD was drilled in the Bay Area, underneath Oakland or Berkeley, along the Hayward Fault, then the 2004 quake could have triggered an unwanted ripple-effect reaction from the public. People might believe—against all “objective” evidence to the contrary–that the USGS had caused the quake by drilling into the fault zone.
The resulting waves of fear and panic could endanger the very existence of seismology as a legitimate government funded science. It’s not too difficult to imagine this happening; and it explained at least in part why the USGS continued to use Parkfield for its earthquake experiments. They needed a guinea pig, a geological lab rat, one that nobody knew or cared too much about; and if we wanted to understand earthquakes and their behavior, we needed to let them conduct their experiments beyond the glare of public scrutiny.
The 2004 Parkfield earthquake might’ve been the most “captured,” mapped, and studied quake ever recorded during the Parkfield Experiment. As such it was probably also one of the most significant earthquakes in the history of seismology and for the future of earthquake prediction; and to my mind it’s at least possible that it may also have been one of the most significant instances of humans directly and measurably impacting the seismic health of our planet. But when I visited six years later few people involved with the USGS wanted to consider this possibility and fewer people in Parkfield even remembered or cared about the quake; and on some level, for the sake of earthquake science and perhaps for the sake of our future survival this was exactly how it had to work. On some level I appreciated the silence, the quiet surface in Parkfield and the peaceful pace of life in the Cholame; and though there was little to see, I found myself returning several more times that summer, a intentional tourist turned loose in the earthquake lab, searching for something big.
 Snyder also told me, “We are well satisfied with all the media coverage (local, regional, national & international – print, radio, TV & documentary) we’ve had over the last few years on the Parkfield Earthquake Experiment and the SAFOD Project. Nothing has changed of late and we’re in the “monitoring phase” for all the data coming in.”