Earth as a Planet: Surface and Interior 209
and the inner core began to grow. The release of the gravita-
tional energy associated with the precipitation of solid iron
is believed to be an important source of the energy driving
the dynamo. Again, estimates are difficult, but models yield
a current temperature range of 5000–7500K.
Seismologically, the inner core has been considered
quite uninteresting, with a very small variation of the phys-
ical parameters across the region. This all changed in
the mid-1980s when it was discovered that this region is
anisotropic, with the symmetry axis roughly parallel to the
rotation axis. A deviation from that symmetry and an ob-
servation of temporal variation of travel times through the
inner core brought forward an interpretation that the in-
ner core rotates at a slightly (1◦/year) higher rate than the
mantle. This is being explained by the electromagnetic cou-
pling with the dynamo field of the other core. However,
this observation has soon become very controversial. Sev-
eral studies now indicate that this differential rotation must
be much less. In 2002, it was proposed that there exists
an “inner-most inner core,” the central region with some
300-km radius in which the anisotropy is distinctly different
than in the bulk of the inner core. Since then, the anoma-
lous properties of this region have been confirmed by other
studies.
7. Earth in Three Dimensions
Figure 21 is an example of results obtained using global
seismic tomography (GST). It shows a triangular cut into an
Earth model of the shear velocity anomalies in the Earth’s
mantle and shows only deviations from the average: if the
Earth were radially symmetric, this picture would be en-
tirely featureless. The surface is the top of the mantle
(Mohorovicic discontinuity, or Moho) and the bottom is the
core–mantle boundary. Seismic wave speeds higher than
average are shown with blue colors, whereas slower than
average are shown as yellow and red colors. Seismic veloc-
ities decrease with increasing temperature: the inference
is that the light areas are hotter than average and dark are
colder. Seismic wave speeds also vary with chemical com-
position, but there are strong indications that the thermal
effect is dominant.
Density is also a function of temperature. Material hot-
ter than average is lighter and, in a viscous Earth, will tend
to float to the surface, whereas colder material is denser
and will tend to sink. Thus our picture can be thought to
represent a snapshot of the temperature pattern in the con-
vecting Earth’s mantle. In particular, the picture implies
a downwelling under the Indian Ocean and an upwelling
originating at the core–mantle boundary under Africa; sec-
tions passing through this anomaly indicate that this up-
welling may continue to the surface. This “window into the
Earth” shows the outer core (blue), inner core (pink) and the
FIGURE 21 A three-dimensional model S362D1 of Gu et al.
representing the lateral deviations of the shear velocities with
respect to PREM. The sides represent a vertical cross section
along three different profiles. Faster than average velocities
(caused by colder than normal temperature, presumably) are
shown in green/blue and slower (hotter) in yellow/red colors.
The scale is±1.5%; significant saturation of the scale occurs in
the upper mantle. Note the lateral and vertical consistency of the
sign of the anomalies over large distances and depths. The
mantle underneath Asia and the Indian Ocean is fast at nearly all
depths, whereas the mantle under central Africa is slow. The
liquid outer core is shown in blue, inner core in red and the
innermost inner core in red.innermost inner core (red); the latter represents only 0.01%
of the Earth’s volume.
The GST is limited by the distribution of globally de-
tected earthquakes and by the locations of seismographic
stations. There is not much that we can do about the distri-
bution of seismicity, except that now and then an earthquake
occurs in an unexpected place, so the coverage is expected
to improve with time. Generally, the earthquake distribu-
tion is more even in the Northern Hemisphere. Much has
been done in the last decade to improve the distribution and
the quality of the seismographic stations, and recent results
show considerably better resolution of the details in the top
200 km, for example. However, even using the available
oceanic islands (which are very noisy, because of the wave
action), there are oceanic areas with dimensions of several
thousand kilometers where no land exists. A series of exper-
iments by Japanese, French, and American seismologists
have demonstrated that the establishment of a permanent
or semipermanent network of ocean bottom high-quality