Geoscience Reference
In-Depth Information
Figure 8.15. Laboratory
experiments with a
moving rigid upper
boundary indicate that
flow in the upper mantle
could take this type of
form. (After Richter and
Parsons (1975).)
aligned in the direction of shear would exist beneath the plates; its upper boundary
layer, not rigidly attached to the plate, would be the thermal boundary layer.
Three-dimensional rectangular laboratory experiments with silicone oil and a
moving, rigid upper boundary have indicated that such a two-scale flow can occur.
Figure 8.15 shows such a convection system. Another laboratory experiment,
which modelled the thermal effect of the subducted lithosphere by cooling one
of the side walls, gave rise to a single, stable, large-aspect-ratio convection cell.
Again these experiments illustrate that large-aspect-ratio cells can be stable; how-
ever, the exact form of instabilities and secondary flow depends on the particular
physical characteristics of the experimental model, its geometry and boundary
conditions.
Isotopic ratios of oceanic basalts are very uniform and are quite different from
those of the bulk Earth, which means that the mantle must be very well mixed.
This is confirmed by numerical models. Figure 8.16 shows a computer model
of mantle convection in a two-dimensional rectangular box. A square patch of
mantle with physical properties identical to those of the rest of the model is
marked, and its deformation and distribution throughout the mantle are traced at
subsequent times. Within several hundred million years, the convective process
is able to mix upper-mantle material thoroughly. This time is short compared
with the half-lives of the measured radioactive isotopes, indicating that upper-
mantle convection should be well able to account for the general uniformity
of isotopic ratios in oceanic basalts. For the upper-mantle model illustrated in
Fig. 8.16,any body smaller than 1000 km is reduced to less than 1 cm thick within
825 Ma.
The isotopic ratios of oceanic-island basalts (OIB) require a source for these
magmas that is less depleted than the source of mid-ocean-ridge basalts (MORB).
Efficient mixing of either the upper mantle or the whole mantle by convection
suggests that the source of OIB must be a recent addition to the mantle or an
unmixed reservoir. If this were not the case, the source would be mixed into the
mantle too well to allow the characteristic isotopic signatures of OIB to have
developed. The source of OIB is thus a matter of considerable conjecture. It is
possible that they originate from the base of the lower mantle and that this is a
Search WWH ::




Custom Search