Geoscience Reference
In-Depth Information
Fig. 20.12
(a) Azimuthal
anisotropy of 200-s Rayleigh waves.
The lines indicate the fast phase
velocity direction. The length of the
lines is proportional to the
anisotropy (Tanimoto and Anderson,
1984). (b) Flow lines at 260 km
depth for the upper-mantle
kinematic flow model of Hager and
O'Connell (1979). The model
includes a low-viscosity channel in
the upper mantle. The flow lines
mimic hotspot tracks, and the
relative motions of hotspots,
suggesting that entrained
asthenospheric heterogeneities
might explain melting anomalies on
plates.
of this map is that the vectors are parallel to
so-called hotspot tracks; if fertile heterogeneities
occur in the asthenosphere and are responsible
for hotspots then these hotspots will appear to be
motionless on a given plate, but will move with
respect to hotspots on other plates. For example,
the hotspot tracks on the Nazca plate will appear
to move relative to those on the Pacific plate, and
hotspots in the Atlantic, Antarctic and Africa will
appear to be stationary.
In the kinematic flow model the flow is nearly
due south under Australia, shifting to southwest
under the eastern Indian Ocean, or directly from
the subduction zones to the nearest ridge. In
the anisotropic map the inferred flow is more
southwestward under Australia, nearly parallel
to the plate motion, shifting to east-west in the
eastern Indian Ocean. The southeastern Indian
Ridge is fast at depth, suggesting that this ridge
segment is shallow. The Mid-Indian Ridge, the
Indian Ocean triple junction, and the Tasman Sea
regions are slow, suggesting deep hot anomalies
in these regions. These deep anomalies are off-
set from those implicit in the kinematic model
and apparently are affecting the direction of the
return flow.
Similarly, the flow under the northern part of
the Nazca plate is diverted to the southwest rela-
tive to that predicted, consistent with the velocity
anomaly observed near the southern part of the
Nazca-Pacific ridge. The flow lines in the mantle
under the Nazca plate are parallel to the hotspot
tracks. The anisotropy due north of India indi-
cates north-south flow, perpendicular to the plate
motion of Eurasia and the theoretical return-flow
direction. One interpretation is that the Indian
plate has subducted beneath the Tibetan plateau
and extends far into the continental interior.
Note that anisotropy and predicted flow vec-
tors are subparallel under the Pacific plate but
