ScienceWeek

EARTH SCIENCE: ON CONVECTION IN EARTH'S MANTLE

The following points are made by Albrecht W. Hofmann (Nature 2003 425:24):

1) Sandwiched between Earth's thin crust and its metallic core lies a layer of pressurized rock at high temperature -- the mantle. Convection in this layer drives plate tectonics and sea-floor spreading, but we know little about the pattern of circulation. Indeed, current thinking about mantle dynamics is in a state of turmoil. As we cannot observe convection directly, we must piece together indirect evidence from seismology, geochemistry, mineral physics, fluid dynamics and numerical simulations of convection. But the evidence is contradictory and has led to at least two conflicting views about mantle movement.

2) The two conflicting models for mantle convection are usually described as "layered" convection (supported by geochemists) and "whole-mantle" convection (supported by seismologists). Geochemists have long insisted on the two-layered model, in which the mantle consists of a relatively primitive layer below a depth of 660 km -- containing primordial noble gases, trapped 4.5 billion years ago when the Earth formed -- and an upper layer that is highly depleted of heat-producing elements (uranium, thorium and potassium), noble gases and other "incompatible" elements. The primitive layer serves as a reservoir for these elements (which were depleted from the upper mantle when Earth's crust was formed) and it is occasionally sampled by deep-mantle plumes.

3) Over the past several years, however, seismic tomography has given us increasingly detailed images of apparently cold "slabs" (characterized by fast seismic velocities) descending into the deep mantle right through the 660-km boundary, effectively cutting to shreds the simple picture of the mantle convecting in two nearly isolated layers. If cold "slabs" descend into the deep mantle, there must be a corresponding upward flow of deep-mantle material to shallow levels. No matter what specific form the exchange across the 660-km boundary takes, in this "whole-mantle" model of mantle convection, it would within a few hundred million years destroy any compositional layering that had possibly been inherited from early in Earth's history.

4) Meanwhile, the geochemical arguments for a separate deep reservoir have not disappeared. Primordial noble gases are still preferentially associated with "hot spots'(2), at least some of which seem to come from deep-mantle plumes(3). And much of the upper mantle remains highly depleted of incompatible trace elements, including the heat-producing thorium, uranium and potassium -- also suggesting the presence of a less depleted reservoir deep within the mantle(1,4,5)

References (abridged):

1. Bercovici, D. & Karato, S. Nature 425, 39-44 (2003)

2. Graham, D. W. Rev. Mineral. Geochem. 47, 247-317 (2002)

3. Montelli, R. et al. Geophys. J. Int. (in the press)

4. Hofmann, A. W. Nature 385, 219-229 (1997)

5. Davaille, A. Nature 402, 756-760 (1999)

Nature http://www.nature.com/nature

ScienceWeek http://scienceweek.com