Geoscience Reference
In-Depth Information
the genera are endemic, that is, found only at hydrothermal vents. Not only do
the mid-ocean ridges act as the highway along which fauna can migrate/disperse
between the hydrothermal vents (even though individual vents and vent fields
are transient features), but also the timings of the past links between mid-ocean
ridges (e.g., Galapagos, Gulf of California, Juan de Fuca Ridge and Mid-Atlantic
Ridge) all seem to be partially reflected in the present-day distribution of fauna.
At subduction zones, the influence of the hydrothermal circulation is again
felt and not only in the hydrothermal vent fields on spreading centres in back-arc
basins. On a totally dry planet, the subducting crust (if subduction took place at
all) would be without water. On Earth, the descending plate includes much water
and CO
2
(even if virtually all the sediments lying on top of the oceanic crust
are scraped off; see Section 10.2.2) because the slab is so heavily hydrated and
carbonated. The water and CO
2
in the descending plate are eventually driven off
and rise into the overlying mantle wedge. There the water plays a critical role in
promoting melting, since wet rocks melt at much lower temperatures than do dry
rocks. This melting (Section 10.2.1)isthe process that leads eventually to the
magmas that have formed the continents. Finally, a very small amount of water
and CO
2
can descend with the plate into the deeper interior and can recirculate
through the upper mantle over billions of years, eventually to rise again with the
mantle under mid-ocean ridges. Even the very small amount of water that exists
in ridge lavas has important geochemical and tectonic effects, such as promoting
the melting process.
9.4.5 Seismic structure
The possibility of the existence of extensive, crustal infinite-onion magma cham-
bers (Fig. 9.20) has led to a number of seismic experiments over the axial regions
of the mid-ocean ridges in an attempt to delineate these chambers. Such magma
chambers should be characterized by low seismic velocities and high attenuation.
Early models of theseismicstructure of the oceanic crust (Figs. 9.11(e)-9.11(g))
had a very wide zone centred on the ridge axis in which normal layer-3 and
upper-mantle velocities were not measured. However, subsequent experiments
have revealed that the crust has the 'normal' oceanic structure except over a very
narrow axial zone (about 20 km wide in some cases). Within this zone, layer 3
often appears to be absent or has a reduced velocity, and normal upper-mantle
velocities are frequently not measured. Instead, the highest velocity is generally
7.1-7.6 km s
−
1
.
Slow-spreading ridges
Seismic experiments shot on the Mid-Atlantic and Reykjanes Ridges have yielded
upper-mantle velocities at the ridge axis of 7.1-7.6 km s
−
1
. The final, well-
constrained seismic-velocity models for one of these experiments, for which
synthetic seismograms were used as part of the modelling procedure, are shown
