Geoscience Reference
In-Depth Information
which seismic waves can be reflected. So perhaps the layers indicate a change in crystal
structure rather than composition.
A double boiler?
The 660-kilometre layer, the division between the upper and lower mantle, is a particularly
strong feature and the focus of vigorous debate between those who think that the entire
mantle is circulating in a huge convection system and those who think that it is more like
a double boiler with separate circulating cells in the upper and lower mantle and little or
no exchange of material between them. Historically, geochemists tend to favour the double
structure as it allows for chemical differences between the layers, whereas geophysicists
prefer whole-mantle convection. Present indications are that both might be right, in a com-
promise solution in which whole-mantle circulation is possible but difficult. Data from
seismic tomography would seem at first to favour the double boiler idea. The seismic scans
reveal where slabs of subducted ocean crust sink down towards the 660 km anomaly. But
they do not seem to pass through it. Rather, the material spreads out and seems to collect
at that depth, for hundreds of millions of years. But further scans show where it can break
through like an avalanche and continue on through the lower mantle almost to the top of
the core.
In June 1994, Bolivia was shaken by a powerful earthquake. It did little damage because
its focus was so deep - about 640 kilometres. But at that depth, the rocks should be too
soft to fracture. This is a region where a slab of old ocean crust from the Pacific is sinking
down beneath the Andes. What must have happened is that a whole layer of rock under-
went a catastrophic phase change into the denser perovskite structure. That seems to be ne-
cessary before it can sink down into the lower mantle. The explanation solves the mysteries
of mantle layering and deep earthquakes at one go.
But there is much that still needs explaining. For example, the slab of ocean crust that is
subducting below the Tonga trench in the Pacific is passing into the mantle at about 250
millimetres per year, far too fast for its temperature to even out. Material would reach the
base of the upper mantle in just 3 million years and its low temperature should be obvious
if it pools there or extends into the lower mantle. But there is no evidence for such a slab.
One theory is that not all of the olivine converts into higher-density minerals, making the
old slab neutrally buoyant in the upper mantle. The combination of cool temperature and
mineral content would give it a seismic velocity very similar to other mantle material, so it
would not show up easily, just as a layer of glycerine does not show up well in water. There
is indeed tantalizing faint seismic evidence for such a slab deep below Fiji.
Search WWH ::




Custom Search