Geoscience Reference
In-Depth Information
The temperature measurements discussed previously suggest that the tempera-
ture assumptions of the second model are too low. When this second model is
recalculated using more appropriate liquidus temperatures of 1230-1250
◦
C and
solidus temperatures of 1160-1185
◦
C, the maximum half-width of the magma
chamber is about 2 km (or less if the latent heat is released entirely at the solidus).
Latent heat of about 4.2
10
5
Jkg
−
1
is a reasonable estimate for basalt.
The effect of accumulation of crystals settling to the bottom of a magma
chamber can also be included in the thermal models. The width of any chamber
is reduced by crystal settling: a 2 : 1 ratio of solidified material forming the
chamber roof to that falling to the bottom reduces the maximum width of the
magma chamber to about one-half its previous value, whereas a 1 : 1 ratio results
in a reduction of width by one-third. Crystal settling also has a significant effect
on the shape of the magma chamber, changing it from triangular (apex at the top)
to a diamond shape.
Thermal models calculated for medium- and fast-spreading ridges are very
similar to those for slow-spreading ridges, the only significant difference being
that the magma chambers are much wider, as one would expect. Thermal con-
siderations therefore imply that a steady-state crustal magma chamber can exist
on fast-spreading ridges, though, for half-spreading rates less than 1 cm yr
−
1
,it
is questionable whether any steady-state crustal magma chamber can exist. For
the Mid-Atlantic Ridge, the maximum half-width is probably 0.5-1.0 km, and
any magma chamber would then be confined to depths equivalent to the lower
part of layer 3 (i.e., 4-6 km below the seafloor). For the East Pacific Rise, a half-
width of 3-5 km would be generally appropriate. Note that, if much hydrothermal
cooling of the crust occurs, then these conductive thermal models are not appli-
cable because cooling would be very much faster. This means that the size of any
magma chambers would be much reduced.
Figure 9.21(b) shows a possible velocity model for a fast-spreading ridge, the
East Pacific Rise at 9
◦
N (half-spreading rate 6.1 cm yr
−
1
). This model was derived
using the first thermal model discussed, P-wave velocities measured on the Semail
ophiolite samples, a rate of change of seismic velocity with temperature d
a
×
/
d
T
10
−
4
km s
−
1
◦
C
−
1
and an arbitrary 1.5 km s
−
1
reduction in seismic veloc-
ity in the magma chamber to model the partial melt (compare with Fig. 9.13).
The maximum width of this magma chamber is about 5 km, and its top is some
2km beneath the seabed.
of
−
8
×
9.4.4 Hydrothermal circulation in young oceanic crust
Hydrothermal circulation through the ocean floor is one of the most important
geochemical and geophysical processes on Earth. It plays a major role in con-
trolling the chemistry of sea water, in the operation of subduction zones, in the
growth of continents and in managing the Earth's heat budget. The loss of heat by
hydothermal circulation is estimated to be one-third of the oceanic heat flux or
