Scientists Concerned as San Andreas Fault Observatory Seeks Additional Funds

SAFOD drilling site
Drilling of the San Andreas Fault Observatory at Depth was in progress when I visited the site in Parkfield, California, in September 2005 (Andrew Alden photo)

It was a truly history-making mission to the deep Earth that drilled a 2-mile borehole to the active trace of the San Andreas fault. Today the San Andreas Fault Observatory at Depth, or SAFOD, languishes just short of its goal, and scientists are scratching for funds to place instruments where they were meant to be.

SAFOD began ten years ago as the boldest part of the great EarthScope research program. Two other EarthScope projects have crossed the nation with a giant movable set of seismometers (USArray) and blanketed the western states with GPS receivers to monitor crustal movements (Plate Boundary Observatory). The SAFOD project gave us the means to witness earthquakes close up, while they happen.

A site was picked in the tiny town of Parkfield, California, where the San Andreas fault is well behaved (for a fault) and where earthquake scientists have had a presence for many years. A scientific drill rig was set up on a site provided by a generous cattle rancher, and work began in 2002 with an exploratory hole. Three years later, I saw the rig in action as it prepared the main borehole, which started straight down, then curved eastward to intersect the fault from the side. As expected, the fault zone itself was marked by about 200 meters of shattered and altered rock.

SAFOD grit
Pulverized rock from the borehole was separated from the drilling mud after each ten meters of progress. Minerals and microfossils in this grit enabled the drillers to map the underground in detail as they went (Andrew Alden)

The next trick was to find the active plane in that wide fault zone. The borehole was lined with steel casing and then monitored for a couple of years. The fault was pinpointed in two places where it was gradually warping the casing. The next phase of the project targeted these places, drilling sidetracks from the main hole and collecting cores of undisturbed rock from them. These alcoves were where a set of six sensitive instruments—the downhole observatory—was installed in September 2008.

SAFOD first core
The first core is retrieved on August 6, 2007. The cores came up hot, around boiling temperature (Clay Hamilton/SAFOD)

The cores, as precious as moon rocks, were displayed to the press on October 5, 2007. It was a privilege to see them with my own eyes. Today they are stored in a Texas facility, where samples are parceled out to qualified researchers.

SAOFD core closeup
Life-size photo of the core from the fault zone. The brown is slippery clay; the green rock is serpentinite (Andrew Alden)

Unfortunately, the downhole instruments failed after a few weeks in the harsh conditions. A new seismometer was lowered into the hole and continues to yield good data; SAFOD scientist Bill Ellsworth told me that a replacement is on the way. But beyond a shoestring of funding that various sources have been scraping together, the money is gone and we're left with half the result that scientists hoped for.


The National Science Foundation, which paid for EarthScope, announced in April that it wouldn't spend any more on SAFOD, ignoring the pleas of a wide set of quake scientists. Earlier this week, the Mercury News reported that the project's scientists are preparing an unsolicited bid for funding to the NSF, on an equal footing with hundreds of other research proposals.

If they succeed, it would be a good thing as SAFOD is not alone. Japan's well-funded NanTroSEIZE program has been drilling across active faults under the seafloor, and the Deep Fault Drilling Project is drilling across a major fault in New Zealand. These days, we could really use fresh data from the San Andreas.