A 2,000-kilometer-long slab of unattached lithosphere may be suspended 400 to 600 kilometers below the Fiji islands near the Tonga subduction zone. It may have broken off from a previous subduction along the now dormant Vitiaz trench 5 to 8 million years ago, which would indicate that mixing in the mantle is less dynamic than supposed and pristine lower mantle more likely to exist.
[Left: A 3-D rendition of a detached slab remnant and the actively subducting lithosphere in the Tonga subduction region constructed from 35 years of seismic records. The juxtaposition of slabs retains a large amount of subducted material above the lower mantle. Image credit: Wang-Ping Chen and Michael Brudzinski]
Wang-Ping Chen and Michael Brudzinski of the University of Illinois, Urbana, presented evidence for the slab in the June 29 Science, adding fuel to a debate over the interaction between subducted lithosphere and the transition zone between upper and lower mantle. “Is the whole mantle churning like soup on the stove or is it moving in layers or what?” says Harold Green of the University of California, Riverside, and author of an article in the same issue of Science evaluating Chen and Brudzinski’s work. “They’ve raised some questions for someone else to chew on.”
Chen and Brudzinski used two pieces of evidence to infer the slab’s presence. One is the deep earthquakes located near, but not in, Tonga’s Wadati-Benioff zone (WBZ), where the cold Pacific plate is plunging at a rapid 2 centimeters per year — five times faster than movement along the San Andreas fault — underneath the Indo-Australian plate. In addition to the usual Benioff quakes, data from the last 35 years show a cluster of earthquakes 300 kilometers deep and several hundred kilometers west of the WBZ. Another group is located at 500 kilometers depth, 200 kilometers above the deepest end of the WBZ.
Earthquakes located deeper than 100 kilometers are assumed to take place in cold subducted lithosphere. But the location and focal mechanism of these quakes indicate that they are not happening in the actively subducting Tonga lithosphere.
The second piece of evidence is the slower-than-expected speed of P and S waves in the region where the unusual earthquakes occur. The speed of seismic waves at the mantle’s transition zone is constrained mainly by temperature and petrology. Subducted material is relatively cold and seismic waves should, therefore, travel faster there than in the surrounding material. Since they don’t, the authors attribute the relatively slow speeds to the presence of either subducted metastable olivine or subducted volatiles, two ingredients that can slow seismic waves and trigger earthquakes. But because not enough earthquakes occur to “map” the slab, the researchers used seismic tomography to outline its orientation and size.
Their findings indicate the presence of a remnant slab lying at a low angle above the active WBZ, which dips steeply at 60 degrees. Although a solitary slab doesn’t leave much evidence of its origin, the fact that it overlaps the WBZ makes it difficult to connect it to the subducting Tonga slab. If it is some kind of remnant from previous subduction activity, then there is hope that the lower mantle may not have been subject to the wholesale penetration of subducted lithosphere, Chen says. “The amount of slab penetration into the lower mantle controls the rate of heat and mass transfer between the upper and lower mantle,” he explains. “Heat is getting out of the earth slower than we thought — it’s not a total convective system. The implications here are that it’s easier to preserve primordial mantle.”
Chen and Brudzinski’s explanation is not the only one for the slab’s origin. Green suggests that the buoyant slab could be a piece of subducting Tonga slab that broke off and rose upward, instead of a piece of older lithosphere slowly sinking. “Where is the material that has been subducted down the Tonga subduction zone?” he writes. “The seismogenic zone in the currently active slab ends at the base of the transition zone (the maximum limit of earthquake activity worldwide).”
Whatever the answer to the unusual seismicity at Tonga, Green points out that a slightly sloping slab is much easier to study from the surface than a steeply sloping slab with no constant depth. And the fact that the unattached slab was found and tentatively “mapped” shows that scientists are making leap-frog advances in their exploration of deep mantle conditions, he says.
Emily D. Johnson