We still don't know when the next big one will hit, but we do have some new information on what may precede it. The findings come from deep below the earth, near the San Andreas Fault in Parkfield.
"Prior to a few earthquakes in the area, we saw a change in the velocity. The velocity went down which says that alright, something's happening, the stress is causing the earth to maybe open up a little bit," says Ernie Majer, a researcher for Lawrence Berkeley National Laboratory.
A pilot study from Rice University and Lawrence Berkeley Labs looked at two small earthquakes near the San Andreas Fault Observatory and measured changes in seismic waves.
"It's really, what we think, a function of the stress causing certain parts of the zone to deform and when it deforms, the elastic properties change, and when those change, the seismic wave velocity will change," says Majer.
The study's authors say this information helps them understand the fundamental mechanisms of an earthquake, which could one day lead to help predicting them, but that day is still a long way off.
"The earthquake community is very hesitant about predicting earthquakes," says Majer.
"It's a curious result, but one that will have to be followed up with much more intensive study," says Bill Ellsworth, a research geophysicist with the USGS.
The USGS says there's still a lot of questions about the findings, but any information on what happens before an earthquake is helpful.
"This study is a good example of the things we can't do at the surface, so it's very encouraging from that standpoint that they're making new measurements and perhaps leading us in new directions," says Ellsworth.
And if those directions do lead to short-term earthquake prediction, there are also questions about that.
But can we predict it? Do you want to predict it? What happens if you said 20 minutes beforehand there's going to be an earthquake? How many people would you kill with people driving away?" says Majer.