A short one. The plume (red-brown) seismically imaged beneath Iceland begins at 660 kilometers.

Mantle Plumes-a-Plenty?

For 3 decades, researchers have been debating whether plumes of hot rock rise through Earth's mantle. Geologists and geochemists have inferred deep plumes from traces left at volcanic hot spots such as Iceland and Hawaii. And some seismologists have suggested that they could glimpse a plume or two in their seismic “CT scans” of the mantle. Now, a team claims to have found evidence of not one or two plumes--but 32. That doesn't settle the debate, though.

The researchers, led by seismologist Raffaella Montelli of Princeton University, introduce a new way of analyzing the seismic waves that are combined into an image of Earth's interior. Usually, seismic waves are considered to follow curved lines called ray paths. Where numerous ray paths traverse hotter than normal rock, the waves are slowed and an anomalously warm spot appears in the image.

But thin, warm structures--such as the supposed plumes--would be particularly difficult to image in the conventional manner. So Montelli and her colleagues developed an analysis that takes into account how seismic waves actually travel, spread across a wave front rather than along a single ray path (Science, 3 January, p. 35). This technique boosted the strength of signals from some plumelike structures by up to 60%, the team reports online today in Science.

The technique shows plumes beneath most classic volcanic hot spots. In addition to the two broad superplumes that everyone sees, beneath Africa and the South Pacific (Science, 9 July 1999, p. 187), the new method also shows narrow plumes rising off them, sometimes splitting before reaching the surface. Elsewhere, lone plumes stretch from near the core-mantle boundary to the surface. A few hot spots, including Yellowstone, seem to lack plumes. And in a major surprise, the plumes beneath two of the most classic of hot spots, Iceland and the Galápagos Islands, begin much closer to the surface than they had appeared to (Science, 14 May 1999, p. 1095).

“We are providing visual proof plumes exist.” says Montelli. Others are more cautious, however. “I think it is fair to at least suspect that they are overinterpreting their data set,” says seismologist Barbara Romanowicz of the University of California, Berkeley. Until several technical questions are resolved, she says, “I think it is a leap of faith to claim a discovery” of dozens of plumes.

Related site
More information about the mantle plume debate