Researchers have pulled up the broken instrument package that has sat dormant deep inside the San Andreas fault for the past few years, to see if they can work out what went wrong and how to fix it.

The San Andreas Fault Observatory at Depth (SAFOD) experiment involved drilling a 3km hole into the ground near Parkfield, California, pulling up the first cores of rock from inside a fault, and installing instruments such as seismometers with the intention of long-term monitoring. The cores have been a great success, but the instrument package only worked for about a month, says Greg Anderson, National Science Foundation (NSF) programme director for the EarthScope programme, of which SAFOD is one part. The shut down was a mystery (see news story). Since then, researchers have pondered whether they would do more harm than good by pulling the instruments back up, and plotted how to do so as safely and cheaply as possible.

Anderson reported at the American Geophysical Union annual meeting that the gear was pulled up a few weeks ago and the instrument package split open. There was no immediately obvious indication of what had gone wrong, Anderson says. The pipe did smell really bad, he told the AGU gathering (the scientist overseeing the extraction yelled out from the audience that he had to throw away his clothes), but engineers assured the team that this isn’t unusual for equipment coming out of a deep hole.

A team is due to report in the spring on whatever they can discover from the broken equipment, along with recommendations for if and how they should try the experiment again.

Meanwhile the other parts of EarthScope are pressing ahead. An array of seismometers that is marching east across the country has reached more than half-way, and should reach the coast by 2012 or 2013.