>
A Solid in Motion
Another important area of research focuses on the rheological behavior of the mantle–in other words, the way in which it continuously deforms. “On our time scale, the mantle appears to be totally solid,” IPGP’s Anne Davaille points out. “But over very long periods of time, we observe that it behaves like a viscous material, and ‘flows’ at a rate of a few centimeters per year. A good analogy would be the stained-glass windows found in cathedrals, whose glass inexorably flows down under the effect of gravity. Of course, the higher the temperature in the mantle, the less resistant it is to flow.” It is this convective motion that cools the planet’s interior. This is because the slow movement of the masses of rock carries the planet’s internal heat towards the surface. This is a much faster and more efficient way of evacuating heat than conduction, another process that plays a role in cooling the mantle. Yet for Xavier Le Pichon, who holds the chair in geodynamics at the Collège de France, things are far more complex. “We still do not know whether the mantle undergoes convection as a whole, or whether two separate levels convect independently on top of each other.”
Recently, Francis Albarède, professor of geochemistry at the École Normale Supérieure in Lyon2, proposed a unifying theory of the fate of rare gases in the Earth that removes the strongest objection against material exchange across the entire mantle.3
Some seismological research points to the existence of another type of convection called “superplumes.” These are believed to be huge mushroom-shaped masses of hot, light matter that form at the core-mantle boundary under the Pacific and under Africa, causing upwelling of hot matter from the D’’ layer.
Be that as it may, convection is the driving force behind plate tectonics and its direct and often destructive effects, including mountain formation, earthquakes, and volcanic eruptions. The Earth’s surface is made up of a patchwork of around a dozen spherical plates, which may be 100% oceanic, 100% continental, or a mixture of the two. The mid-ocean ridges, the area where two plates are moving away from each other, have a total length of 67,000 kilometers, and it is here that approximately 90% of all the magma that flows out from the mantle is released. These spreading zones raise a number of questions. “For instance, we don’t know precisely how the magmatic liquid makes its way up to the surface,” says IPGP’s Mathilde Cannat, also a member of the MoMAR project, which focuses on the Mid-Atlantic Ridge south of the Azores. “We’d like to know if the magma travels along channels, and if so, what size they are. Apart from that, we still have to understand how magma gets onto the ocean floor, the frequency of eruptions, etc.”
Then there are subduction zones, which act as global “differentiation factories,” the places where the lithosphere (the rigid 100 kilometer-thick outer layer of the Earth made up of the crust and upper part of the mantle) continuously descends into the mantle to be recycled. Albarède compares the descent of the plate back into the mantle to a tablecloth that slides off a table because it’s too heavy. “The oldest, and therefore thickest, part of the crust–which was formed directly above the ocean ridges by rising basaltic lava–bends and sinks down into the deeper parts of the Earth together with the sediments on top of it.” If the underlying principles are well known, many questions still remain. Why do some plates sink right down to the base of the mantle while others remain stuck at a depth of between 400 and 670 kilometers? How long does it take for a plate to reach the depths of the mantle before blending with the ambient material and eventually rising back up towards the upper layers?
Intense Activity
To try to better understand the processes that occur at subduction zone, LGCA’s4 Catherine Chauvel focuses her research on the Lesser Antilles island arc, where the Atlantic plate dives under the Caribbean plate, causing intense volcanic (Mount Pelée, La Soufrière, etc.) and seismic activity. Because it is known that volcanic rocks “more or less directly record the processes active in the mantle ‘wedge’–the place where the plate descends–and the conditions prevailing in the subducted plate,” Chauvel explains, “we establish the chemical budget of the material entering the subduction zone, model the variations in the types and composition of volcanic rocks, and understand how transfers of fluids from the descending plate into the mantle wedge occur.”
While spreading zones and subduction zones produce volcanism directly related to plate tectonics, this is not the case for “hot spot volcanism.” It appears to be associated with the D” layer, and acts as a blowtorch beneath the moving plates. Huge plumes of hot material rise up from the core-mantle boundary like giant hot-air balloons, and spread out forming bulges beneath the plates. The pressure drop triggers partial melting and the production of spectacular amounts of basalt. “The gradual build-up of these reservoirs leads to the formation of millions of cubic kilometers of flood basalt, like the Deccan Traps in India,5 and this in a very short period of time, geologically speaking (a million years),”explains Pierre Schiano, director of the LMV,6 one of the world’s largest laboratories dedicated to volcanic activity. “On continents, these increasingly pronounced bulges can even lead to a plate rupture and the opening of a new ocean.” These shallower hot spots may form at the interface between the upper and lower mantle. For now, this extremely potent type of volcanism remains highly mysterious.
Philippe Testard-Vaillant
Left: A volcanic rock that contains peridotite (light green), a material that originates in the upper mantle. Above: A sample of perovskite, a material that is present everywhere in the lower mantle.
© H. Raguet/ CNRS Photothèque
© Antoine Dagan for le Journal du CNRS/CNRS Photothèque
Mass Extinctions 1. Gigantic continental flood basalts. |
Pockets of magma 1. S. Labrosse, et al. “A crystallizing dense magma ocean at the base of the Earth’s mantle,” Nature 2007. 450: 866-9. |
1. Institut de physique du globe de Paris (CNRS / Universités Paris-VI, VII, et de la Réunion).
2. CNRS / École normale supérieure de Lyon / Université Claude Bernard Lyon-I.
3. F. Albarède. “Rogue mantle He and Ne.” Science 2008. 319: 943-5.
4. Laboratoire de géodynamique des chaînes alpines (CNRS / Université Grenoble-I / Université Chambéry).
5. A massive stack of lava flows that form stepped cliffs in Southern India.
6. Laboratoire magmas et volcans (CNRS / Université Clermont-Ferrand-II / Université Saint-Étienne / IRD).
Éléonore Stutzmann,
stutz@ipgp.jussieu.fr
Anne Davaille,
davaille@fast.u-psud.fr
Xavier Le Pichon,
lepichon@cdf.u-3mrs.fr
Mathilde Cannat,
cannat@ipgp.jussieu.fr
Francis Albarède,
albarede@ens-lyon.fr
Catherine Chauvel,
catherine.chauvel@ujf-grenoble.fr
Pierre Schiano,
p.schiano@opgc.univ-bpclermont.fr