As a boy growing up in Illinois, Matt Pritchard's curiosity about geology was sparked by a rock collection he kept as a 4-H project. The geology of the Midwest was less than inspiring, but a family trip that included the Grand Canyon, Yellowstone and Walnut Canyon National Monument in Arizona solidified his interest.

These days, the assistant professor of earth and atmospheric sciences is most familiar with the rocks and ground that move in response to volcanoes, earthquakes and landslides. Using satellites hundreds of miles in space, he can calculate centimeter-scale movements of Earth's surface.

"Now we have these very dense maps of ground deformation from earthquakes," says Pritchard. "Being able to do it from space is a huge revolution."

To get such a precise picture of Earth's movements, Pritchard uses interferometric synthetic aperture radar. InSAR, as it's called, is an unintended byproduct of SAR, which uses microwaves to produce images of Earth, even in darkness or through clouds. Instead of using the amplitude of the wave to produce an image, interferometry uses the phase, which is the position of the wave within its cycle -- trough, peak or somewhere in between. As long as Earth's surface hasn't moved, the phase of a returning wave should be the same each time an InSAR satellite returns to a location. Because the length of the wave is known, a phase shift can be used to calculate the change in the distance the wave has traveled.

In reality, of course, satellites rarely return to the exact same location, and that changes the phase in a couple of ways, but this can be compensated for using the known difference in satellite position and accurate topographic information. Vegetation and plowing, for example, can also distort the returning signal. But when the "noise" is low or filtered out, Pritchard is left with an interferogram. A fringe pattern of alternating red and green bands indicates a change in the surface, like topographic lines on a map, except each band represents just 3 centimeters.

"It's so fantastic," says Pritchard. "It's just a lot of fun every time one of these fringe patterns comes up; it's such a rush."

While a graduate student at the California Institute of Technology, Pritchard worked with scientists at NASA's Jet Propulsion Laboratory to write the open-source software most commonly used to look at InSAR data. With this new tool, he and his Caltech adviser, Mark Simons, discovered that earthquake faults sometimes slip slowly, deep underground, without any violent shaking. Others have found such "silent earthquakes" occur about every 14 months in the Pacific Northwest. Eventually this stress overcomes the force that holds the plates locked and they slip suddenly. That's when the earth trembles and buildings fall.

Although silent quakes have been documented before several giant temblors, not every silent earthquake leads to a big one. Pritchard hopes InSAR will reveal more about the relationship between the two. "We're trying to understand where these silent earthquakes occur and what happens before and after," he says.

Precise earthquake prediction may not be possible, says Pritchard, but accurate earthquake forecasting may be. The larger timeframe of such a warning wouldn't allow mass evacuations, but it would let officials know where to focus preparedness efforts. "In theory, with this method, we can take measurements everywhere, all the time, to figure out if there are any precursory signals," he says.

InSAR can also show bulges or dips caused by magma deep underground. Pritchard and Simons studied 900 volcanoes in the South American Andes, where it would have taken years using GPS surveys, but in just a couple of weeks they found about 40 fewer active volcanoes than were previously believed to exist. That allows scientists to concentrate resources where they are most needed. They also found four volcanoes that had caused ground deformation that were not on the list of potentially active volcanoes, including one that has been "sleeping" for many millennia. Pritchard has found that for the past 15 years it has been causing the ground to inflate about 1 to 2 centimeters per year. "Is this a super volcano accumulating magma for the last 300,000 years," he asks, "or is it a benign accumulation of a granite body?"

Pritchard has proposed more detailed field investigations that might help answer this question. He's also beginning to study some more mysterious movements revealed by InSAR, such as the sharp deformation at the edge of a salt flat known as the Salar de Atacama that showed up unexpectedly in his study of Andean volcanoes and earthquakes and ground outside Seattle that rises and sinks seasonally to see if both are due to groundwater movement or something more ominous. Other applications include studying melting glaciers and monitoring mines that have collapsed, causing huge sinkholes.

"Everywhere you look, there's something deforming," he says. "It's been surprising to a lot of people what we've found."

Robert Emro is a writer with the College of Engineering's Office of Communications and Media Relations. This article is excerpted from Cornell Engineering Magazine.