Geoscience Reference
In-Depth Information
The ridge-push force would then be about 4
×
10
11
Nm
−
1
and the slab-pull
10
12
Nm
−
1
(from Eqs. (8.38) and (8.39)). Equating driving and resis-
tive forces enables estimates of viscosity and plate velocity to be made. Plate
tectonics could operate very effectively over an upper mantle with dynamic
viscosity 10
18
Pa s. Velocities could have been high, about 50 cm yr
−
1
. High plate
velocities may have been necessary during the Archaean in order to maintain a
high rate of heat loss through the oceans since, despite the higher temperatures
and heat generation prevalent at that time, the thermal gradients determined from
Archaean continental metamorphic rocks are relatively low. This topic is dis-
cussed further in Section 10.5.
8
×
8.3 The core
8.3.1 Temperatures in the core
Attempts to calculate the temperature at the centre of the Earth using conduction
models (Section 7.4)fail because heat is primarily convected through much of
the Earth. The fine detail of the temperature structure of the mantle depends on its
dynamic structure. Figure 7.16(a) shows two possible temperature models, one
with the upper and lower mantle convecting separately and the other for the whole
mantle convecting with no boundary at 670 km depth. The temperature struc-
ture of the core is another important constraint on the temperature structure of
the mantle because it controls the amount of heat crossing the core-mantle bound-
ary. Conversely, to calculate the temperatures in the core, it is necessary to start
with a temperature for the base of the mantle. Over 20% of the heat lost from
the Earth's surface may originate from the core. This means that the core has an
important role in mantle convection and plate tectonics. Since the surface area
of the core is about one-quarter of the Earth's surface area, the heat flow across
the CMB is comparable to that at the Earth's surface (Table 7.3). The other major
unknowns are the physical properties, at very high temperatures and pressures,
of the iron and iron alloys of which the core is composed (see Section 8.1.5).
High-pressure melting experiments for iron alloys show that the presence of sul-
phur lowers the melting temperature of iron, whereas oxygen seems to raise it.
Nickel is presumed to lower the melting temperature. Thus the details of the com-
position of the core affect its temperature. Nevertheless, despite these difficulties
core temperatures can be estimated, albeit subject to large errors.
Diamond-anvil laboratory equipment that allows material to be studied at the
very high temperatures and pressures of the core has recently been developed.
Experiments involving diamond anvils differ from the shock-wave experiments
in that they allow samples under study to be maintained at core temperatures and
pressures. Pressures up to 150 GPa are attainable. The pressure at the core-mantle
boundary is about 136 GPa (1.36 million times atmospheric pressure), whereas
the pressure at the centre of the Earth is about 362 GPa (see Section 8.1.2). In
