NEWSFOCUS Down to work. Seismologists are continually trans- planting their subterranean seismometers to paint a seismic image of the deep Earth. America. USArray involves three kinds of seismic networks: a Reference Network of 100 seismometers permanently installed 300 kilometers apart in a loose grid across the lower 48 states; a Flexible Array of 446 seis- mometers that are typically placed 10 kilo- meters or so apart for a few months or years to study a feature of particular interest; and the novel Transportable Array, an 800-kilometer-wide net of 400 advanced seismometers 70 kilometers apart. The novelty of the Transportable Array is its combination of broad coverage, relatively dense instrument spacing, and mobility. The GEOPHYSICS array started out hard against the West Coast in 2004 and has been steadily creeping eastward. Today its net spreads 2000 kilometers along Scoping Out Unseen Forces the Rocky Mountains from the Canadian border to the Mexican border. Each month, Shaping North America about 18 instruments on the west side of the array that have collected a couple of years’ As it sweeps across America, the USArray network of seismometers is revealing an worth of data are removed from their 2-meter- on September 25, 2009 REPRINTED BY PERMISSION FROM MACMILLAN PUBLISHERS LTD. impressive but often befuddling subsurface menagerie of slabs, drips, and plumes deep vaults and reinstalled on the east side. Reusing the equipment keeps the project Unlike geologists, who can reach only a few now early in its second 5-year run. Earth- affordable. Over the course of 10 or 12 years, kilometers below Earth’s surface, geophysi- Scope’s three-pronged approach is creating an the Transportable Array will occupy 1600 cists routinely probe thousands of kilometers evolving three-dimensional picture of the locations from coast to coast. Since 2004, all , 439-444 (24 MAY 2009), , 439-444 (24 MAY down in search of the ultimate forces that cre- North American continent. In one component, of USArray has generated 14.3 terabytes of 2 ated and still shape the ground we tread. But researchers drilled through the San Andreas data, nearly as much as the Global Seismo- so far, geophysicists’ picture of Earth’s inte- fault (Science, 12 October 2007, p. 183). In graphic Network has produced since 1988. www.sciencemag.org rior has been maddeningly fuzzy. To sharpen the second, they are gauging the changing The more data collected and the more it, they are scanning the deep subsurface as strain on the crust as it is deformed by deep closely spaced the instruments, the sharper NATURE GEOSCIENCE NATURE never before, pushing a fly’s-eye-like network stirrings and jostling tectonic plates. the pictures of the interior. The most heavily , of seismometers across the lower 48 U.S. EarthScope’s third component—the used seismic imaging technique—seismic states. Researchers “are jumping up and $13.6-million-a-year USArray program— tomography—works like a computed tomog- ET AL. down” with all the new data, says seismolo- looks much deeper. The USArray system raphy (CT) scan of the human body. In a CT gist Edward Garnero of Arizona State Univer- records seismic waves from distant earth- scan, different body parts absorb x-rays to sity (ASU), Tempe. “We’re pretty ecstatic.” quakes after they’ve passed through—and different extents; in seismic tomography, it is Downloaded from And sometimes they’re pretty bewildered. been altered by—the rock beneath North rock’s varying effect on the velocity of seis- “There are so many [imaged] structures under mic waves that paints the picture. the western U.S.,” says seismologist Eugene Waves pass through colder rock Humphreys of the University of Oregon, faster, for example—yielding a patch Eugene. It’s like “we just wandered into a of blue in tomographic images—and dark room and someone turned on the lights. through hotter rock more slowly, ren- We’re struggling to make sense of it.” dered as red. Clearly, the great blobs and chunks of rock rising, sinking, or just floating beneath the A deep zoo surface bear some relation to overlying For the first time, seismic tomogra- mountains, basins, and volcanic outpourings, phers are incorporating substantial but even the avalanche of new data can’t amounts of USArray data into images always resolve exactly what the imaged fea- of the deep western United States. tures are or how they are shaping the surface. Already, the new images have added fuel to a long-running debate over the A creepy-crawler camera existence of mantle plumes (Science, The data surge comes courtesy of the What a drip! Seismic waves that are slower or faster than normal 22 September 2006, p. 1726). One U.S. National Science Foundation’s (NSF’s) (blues or reds, top) can create a 3D image (blue, bottom) of a contingent of researchers studying $25-million-a-year EarthScope program, sinking “drip” tilted by “blowing” mantle rock (dashed arrows). tomographic images had seen these FIGURE ELEMENTS FROM J. D. WEST OF THE IRIS CONSORTIUM; CREDITS (TOP TO BOTTOM): COURTESY 1620 25 SEPTEMBER 2009 VOL 325 SCIENCE www.sciencemag.org Published by AAAS NEWSFOCUS tall columns of hot rock rising The flow of the Great Basin Drip thousands of kilometers from tugging on the crust would explain deep in the lower mantle like a mysterious patch of crust under smoke from a stack. Where compression amid the Great plumes reach the surface, Basin’s pervasive crustal exten- those researchers say, the ris- sion, the group says, although oth- ing hot rock melts and feeds ers see mantle flowing around a hot spots like the volcanoes of slab fragment rather than a drip. Hawaii or Iceland or the geysers and boiling The Aspen Anomaly, a stretch pools of Yellowstone. But other scientists saw of rock that slows seismic waves hot rock extending no deeper than a few hun- dramatically, sits directly beneath 80% of dred kilometers and considered hot spots the Colorado’s 14,000-foot-(5100-meter)- products of tectonic plate interactions, not and-higher peaks as well as the ore-rich heat from the deep interior. Colorado Mineral Belt. Researchers pre- The putative plume beneath Yellowstone sume there’s a connection between the was among the most suspect of some 30 pro- anomaly and the mineralized uplift, but it posed plumes (Science, 3 January 2003, p. 35). remains unproven. And the High Lava But with USArray it’s coming back. Tomogra- Plains of southeastern and central Oregon— pher Richard Allen of the University of Cali- the world’s largest volcanic province of the fornia, Berkeley, and colleagues reported at past few million years—must be guarding the last December’s meeting of the American Geo- secret of their origins somewhere beneath physical Union (AGU) that Trans- them in a mix of sinking slab fragments, a pos- portable Array data add to evi- sible plume tail, and flowing mantle rock dence of a seismically slow zone Eastward ho. The 400 seis- that’s showing up in the latest data. beneath Yellowstone extending to mometers of the Transportable a depth of at least 1000 kilometers. Array will soon move out of the All together now on September 25, 2009 “The whole history of mantle West and reach the Atlantic At the midpoint of the Transportable Array’s plumes makes you hesitate,” says Ocean by 2013. cross-country march, researchers wish USAr- Allen. Still, he says, “I feel pretty ray were yielding more insights and prompt- confident about a plume to lower mantle mic waves pass through it unusually fast, ing less squabbling. “We’re getting a clearer depths.” Unlike the bolt-upright columns geo- prompting speculation that it is denser due to vision in the West,” says van der Hilst, but scientists imagined when they first conceived the composition of its rock. That higher den- “when you look at the details, people do see of plumes in the 1970s, Allen says, his group’s sity might have made it fall away (or drip different things. The [tomographic] models Yellowstone plume slants to the northwest away, as geophysicists say) from the base of allow for different interpretations.” Fouch www.sciencemag.org through the upper mantle and balloons into a the Sierra Nevada. Relieved of that burden, notes that with each group’s different process- much broader slow zone below 660 kilometers the less dense crust could have floated up to ing of the same data, “you can let tomography in the lower mantle. It even seems to have torn form high mountains. become a Rorschach test.” the cold slab of oceanic plate sinking eastward In part to test the Sierra Nevada drip idea, Researchers say they’ll soon find better through the upper mantle under the continent. seismologists led by George Zandt of the Uni- ways to interpret USArray observations. “It’s Other researchers, however, see different versity of Arizona, Tucson, superimposed the such an unwieldy mass of data,” says geophysi- pictures. Seismologist Matthew Fouch of Flexible Array on the Transportable Array as it cist Craig Jones of CU Boulder. “Playing with ASU Tempe agrees that there’s “no clear evi- was passing over the Isabella Anomaly. The it is a different game than we’re used to. I have Downloaded from dence of a simple mantle plume” beneath sharpened view showed a narrower anomaly a feeling we’ll be seeing in the next 5 years Yellowstone. Rather than a contorted colum- than before, which allowed the group to calcu- analyses far more imaginative than what we’ve nar plume, Fouch and colleagues say, their late a density for the anomaly’s rock.