The 10,000-foot peaks in western Oregon known as the Three Sisters were once a flat stretch of real estate. Then the ground began to tremble and bulge, and the first in a long series of volcanic eruptions spewed ash and debris as far south as California and out into the Pacific Ocean. Such eruptions have been building the Sisters from lava and cinders for about 700,000 years. The last one occurred 2,000 years ago, at South Sister; it's been pretty quiet ever since.

These days all the action is west of the Three Sisters Range in a wilderness area distinguished only by its formidable mosquitoes. There the Earth's crust is heaving up in a bull's-eye pattern 10 miles wide. At its center is a four-inch rise, a veritable molehill compared with the nearby mountains. But geologists think it could be the start of something big: the birth of another volcano.

"We really don't know what to make of it yet," says Charles Wicks of the U.S. Geological Survey in Menlo Park, California, whose team found the uplift last May by comparing satellite images taken in 1996 and 2000. Wicks and his peers believe the Oregon event may provide the first opportunity to track a volcanic eruption from its inception, long before the smoke and pyrotechnics begin.

The uplift was most likely caused by a sudden influx of molten rock more than four miles below the surface, says Wicks. That thick, liquid rock, called magma, melts deep within the Earth and then floats upward into fissures in the Earth's crust, where it pools in underground reservoirs called magma chambers. When the pressure in a chamber gets high enough, the magma pushes through existing cracks in the Earth's surface or creates new ones. The eruptions can be slow and deliberate or sudden and catastrophic; both kinds have occurred countless times in the turbulent past of the Three Sisters Range.


Satellite photos taken in 1996 and 2000 reveal a small upwelling stretching over 10 miles. The upwelling rises four inches at its center. Geologists believe that four miles beneath the upwelling's surface lies an enormous chamber of molten lava.
Illustration courtesy of USGS/Charles Wicks

Fortunately, magma upwelling often produces telltale signs that help researchers forecast eruptions and protect human life. Swarms of small earthquakes occur, for instance, when the pressure of rising magma becomes forceful enough to crack rock. The ground above a supercharged magma reservoir may get pushed up by more than two feet. Magma upwelling releases carbon dioxide and other gases from liquid rock, and these gases can be detected at the surface. Escalation of any of these factors—seismic activity, ground deformation, or gas anomalies—may portend an eruption.

But the timeline for such escalation varies from hours to months. When Mount Saint Helens in southern Washington erupted in 1980, seismic activity and surface deformation went from dormant to ballistic in less than two months. Yet for two decades the Mammoth Lakes region of California has experienced earthquake swarms, ground deformation, and carbon dioxide levels high enough to kill trees without volcanic consequences.

Volcanologists believe that snapshots of incipient volcanoes will help them refine their predictive powers. There's only one problem: Baby volcanoes aren't quite as conspicuous as full-grown ones. No one knows what clues to look for in the years or decades or centuries before the sparks start to fly.

No one knew how to look, either, until recently. In the past decade, geologists have figured out how to use satellite-borne radar imaging, called InSAR, to spot subtle deformations on the Earth's surface. By comparing images of the same place taken at different times, they can detect changes as small as one inch in the elevation of the ground. Terrestrial instruments can track ground deformation too. The problem is that such instruments are often installed only after an active volcano has made its intentions clear. They're rarely in place at the site of dormant volcanoes such as the Three Sisters, let alone in areas with no eruptive history, like the adjacent wilderness. And ground-based instruments measure only deformation at isolated points; they don't give a 3-D picture that helps scientists model what's happening in magma chambers miles below.

Now geologists hope to use InSAR to pinpoint and monitor potential hot spots earlier in the volcanic progression. Because of technical limitations, InSAR's coverage is far from global, and the Oregon uplift is the first ominous-looking site found on American soil. Thus survey scientists spent much of last summer in the Three Sisters Wilderness collecting water from backcountry springs, sampling carbon dioxide levels by helicopter, and airlifting instruments that can transmit data on ground deformation and seismic activity continuously.

So far the monitors have detected just one earthquake: a magnitude 1.8 tremor in August that originated three miles underground at the eastern boundary of the bull's-eye. Survey scientists say residents of Bend and skiers at Mount Bachelor, the nearest populous areas, aren't in any danger yet. "It's hard to say what the significance of that one earthquake is," says Willie Scott of the survey's Cascades Volcano Observatory in Vancouver, Washington. "That sort of thing might be a very important precursor that we never had a chance to detect before because we didn't have a system in place to detect it." Or it could be so common it means nothing, he says.

Carbon dioxide levels over the bulge are higher than expected, adds Scott, and the concentration of chloride and sulfate in springs and streams is well above normal levels as well. Both elements escape rising magma as gases before dissolving in groundwater. Survey scientist Steve Ingebritsen first discovered these chemical anomalies in 1990, while evaluating geothermal activity in the area. He was never able to pinpoint the underlying cause. "The area got a lot more interesting when Chuck Wicks found out about this deformation," says Ingebritsen. "It's kind of neat to see such subtle signs of unrest."

The size of the magma chamber hints that the eruption could be one tenth to one twentieth the size of the Mount Saint Helens event. Scott and Wicks hope to get more satellite images as well as ground-based data to help model the movement of magma and the structure of the subsurface rock beneath the wilderness. Though the deformation they see has a well-defined center, a new volcano might not sprout there. Instead, says Scott, the magma could get squeezed eastward along underground pipes and fissures to surface at one or more of the region's existing lava vents. Or it could end up reviving one of the three dormant Sisters. "We know from other volcanoes that the magma doesn't always come straight up," says Scott. "There can be complications in the plumbing."

When the magma does surface, he says, there's no telling how violent the eruption will be. It could ooze out of a vent with little fanfare, or explode in clouds of ash, cinder, and fire. "It's really a crapshoot: if it does erupt, where it'll be, when it'll be, and what it'll be," says Scott. Only one thing's for sure. This new sister has lost the element of surprise.









The U.S. Geological Survey volcano site details current research: volcanoes.usgs.gov.

The Smithsonian Institution maintains the Global Volcanism Program site. Click on the world map, and find out about any active volcano in the highlighted region: www.nmnh.si.edu/gvp/index.htm.