Composition of the Crust and the Mantle - Quantitative Plate Tectonics - page 14

The limit between crust and lithospheric
mantle is shallower than the thermo-mechanical
boundary discussed above, and is represented by
a sharp surface of discontinuity in the seismic
velocities. Across this discontinuity, the P -wave
velocity jumps from 7.0-7.2 to 8.1 km s 1 , while
the S -wave velocity changes from an average 3.9
to 4.5 km s 1 . This surface is called Mohorovicic
discontinuity, or simply Moho .Itmarksthe
transition from the basaltic rocks of the oceanic
crust, or from the granulite facies of the lower
continental crust, to the upper mantle peridotites .
An example of mineral composition of peridotite
rocks in the mantle lithosphere is: 71.6 % Ol,
23.7 % Opx, 3.6 % Cpx, and 1.1 % Sp (Eggins
et al. 1998 ). In terms of oxide components, these
ultramafic rocks may include: 44.1 % SiO 2 ,
44.5 % MgO, 7.9 % FeO T ,1.3%Al 2 O 3 ,0.9%
CaO, and 0.5 % Cr 2 O 3 . Although the chemical
compositions of the mantle lithosphere and the
asthenosphere are similar, there is a substantial
difference in the degree of fertility, hydration,
and presence of melt. When fertile and wet
asthenosphere melts at a spreading ridge by
adiabatic decompression (Fig. 1.5 ), the residual
column of asthenospheric material leaving the
melting regime is depleted in incompatible
elements and dehydrated. Furthermore, it may
contain retained melts in the lower part, at the
wet solidus boundary. While this column is
dragged horizontally by the overlying oceanic
crust, it is also subject to cooling by conductive
loss of heat through the Earth's surface. The
details of this process will be given in Chap. 12 .
For the moment, it is sufficient to observe that
at any time we can divide the column into an
upper part, where the potential temperature is
fallen below the asthenosphere T p ( 1,280 ı C,
Fig. 1.4 ), and a lower hotter zone, which has
not yet lost a significant amount of heat. The
potential temperature at any depth z within this
conductive thermal boundary layer (TBL) can
be calculated using Eq. ( 1.11 ), starting from
the effective temperature T ( z ).Thebaseofthe
TBL coincides with the lower boundary of the
thermal lithosphere , and is generally used to
define the lithosphere - asthenosphere boundary
(LAB) also beneath the continents, although it

must be emphasized that the material between
the 650 ı C isotherm and the LAB behaves as a
fluid and lacks any elastic or plastic strength.
Although the definition of TBL does not ex-
plicitly mention the chemical nature of the mate-
rial within this layer, there is a substantial differ-
ence between the sub-continental and the oceanic
mantle lithospheres. A first important distinction
is that the first one most likely formed during the
Archean (3.0-3.5 Ga) together with the overlying
crust (Carlson et al. 2005 ), whereas new oceanic
mantle lithosphere is continuously created by
the conductive cooling of residual asthenospheric
columns. A second distinction concerns the TBL
height. The thickness of an oceanic TBL within a
residual column increases progressively at the ex-
penses of underlying asthenosphere, as it moves
away from the ridge and cools. In Chap. 12 , we
shall show that a simple relation exists between
age of a residual column and thickness of the
TBL, which attains 125 km after 80 Myrs. Con-
versely, there is no simple relationship between
age and thickness of the TBL in the case of
the sub-continental mantle lithosphere. Here the
depth to the LAB shows considerable variability
and might reach 300 km beneath the Precam-
brian shields (Fig. 1.9 ).
Figure 1.9 shows a global estimate of the
depth to the LAB, based on seismic anisotropy
data (Plomerová et al. 2002 ). In this kind of
studies, the starting point is the observation that
the crystallographic axes of olivine and orthopy-
roxene aggregates acquire non-random orienta-
tions (lattice-preferred orientations, or LPO) in
response to shear deformation. For example, the
orientations of the crystallographic axes of miner-
als in the asthenosphere will be determined by the
direction of the present day flows within this layer
(Tanimoto and Anderson 1984 ). In particular, the
a -axes will cluster about the flow direction, the
a -axes and c -axes will concentrate in the flow
plane, while the b -axes tend to be aligned with the
normal to the plane of flow. This behavior clearly
determines anisotropy in the propagation of seis-
mic waves (Chap. 9 ) , thereby the wavefront ve-
locity will be different along different directions
of propagation. Such anisotropy can be detected
through a variety of techniques. In any case, the