Monday, March 3, 2008

Continuous GPS monitoring: do you know where you're going?

Honestly, I meant to answer Silver Fox's question about GPS monitoring of deformation across the Basin & Range.* But I got distracted by another commenter, who pointed me to the EarthScope GPS site.

Some background: this isn't the handheld GPS unit that you can use while hiking, and which is good for locating an outcrop (or, in my case, my Subaru). These are high-precision units, and they measure changes in location well enough to monitor plate motions. That alone impresses me - it's confirmation of a theory that we accepted based on very indirect evidence. But GPS can do more - it can measure the quiet deformation that follows a major earthquake**, or swelling of a volcano that's about to erupt, or the deformation that occurs within a continent. That means that it (along with other techniques for measuring active deformation) has the potential to test ideas about the behavior of plates, on their boundaries and within their not-always-rigid interiors.

There's a big geophysics experiment going on right now in North America: the EarthScope program. Among other things, it has set up an array of continuous GPS receivers across the US.

And you can look at the results on Google Earth. (Click on images to see a larger version.)

Each red arrow represents the velocity of an individual GPS station. The longer the arrow, the faster it's moving. The Bay Area, for instance, is moving northwest at about 50 mm/year. Sacramento, on the other hand, is moving more slowly, and more toward the west.

But it gets better. You can also look at the 3D velocities:

Red Mountain Pass, not too far north of me, appears to be sinking, as well as moving toward the west. Not very fast - maybe 5 mm/year, with an error of about half that. (There are error ellipses that you can turn off and on in Google Earth.)

Mt. St. Helens is also dropping (as one might expect, given that the magma's coming out right now):

It looks like Mt. St. Helens is spreading a little, too - the north side is moving north, and the south side is moving south.

And yes, to go back to the previous question, there is a GPS station just south of Wells, Nevada.

Wells is moving west and a little up. The adjacent sites had very similar motion.

Now, this doesn't show changes in rate before and after the earthquake. That, presumably, would require more analysis of the data. And then it would take modeling to figure out whether the fault behaved as one might expect for a normal fault in eastern Nevada.

So no great insights about the Wells earthquake. But I know where I'm going now.

(The .kmz files are available from UNAVCO, if you are interested in exploring for yourself.)

*Thatcher et al., 1999, Science, DOI: 10.1126/science.283.5408.1714.

**Subarya et al., 2006, Nature, doi:10.1038.


Silver Fox said...

Wow, indeed! This is really good information and great images. And thanks for the links.

Silver Fox said...

Some kind of subject - verb mismatch in the previous? Oh, well.

Kim said...

The best thing about it, I think, is the way Google Earth makes it possible to look at the vectors from all directions (well, except from underneath), and the way they are shown with the landscape. It makes it much easier to see what's going on in three dimensions.

Ron Schott said...


I've only glanced at it, but I don't recall seeing what the frame of reference is. What is the fixed reference frame that the plates are moving relative to? Greenwich, England?

Kim said...

I didn't find anything about the frame of reference, either. (Though I didn't follow every link, and I got kind of distracted when I realized that the kmz files showed the vectors in 3D...)