Geology Reference
In-Depth Information
since a small amount of water (less than 0.1%) can substantially reduce the solidus
temperature, at which melting first occurs. It is also true that hydrated forms of
minerals are generally less dense than their dry counterparts, which could provide
the buoyancy required to explain hotspot swells. The effect on density needs to
be better quantified, and it would need to be shown that observed water contents
of hotspot volcanics are consistent with the amounts required to explain the buoy-
ancy. It needs also to be shown that sufficient melt can be produced to explain
the observed volcanism, since, although water reduces the solidus temperature,
substantial degrees of melting still do not occur until the dry solidus temperature
is approached.
However, a remaining difficulty would still be to explain the duration of long-
lived volcanic centres like Hawaii. While a hydrated portion of the mantle, perhaps
old subducted oceanic crust, might produce a burst of volcanism, there is no
explanation offered for how the source might persist for 100 Ma or more. It is
useful to estimate the volume of mantle required to supply the Hawaiian plume
for 100 Ma. The total volume erupted into the Hawaiian and Emperor seamounts
over 90 Ma is about 10
6
km
3
. If we assume that there was about 5% melting of the
source, this requires a source volume of 2
10
7
km
3
, equivalent to a sphere of
diameter 340 km. If such a large and buoyant region existed as a unit in the mantle,
it would rise and produce a burst of volcanism. To explain the Hawaiian volcanic
chain, the hydrated mantle material needs to be supplied at a small and steady rate.
The advantage of the thermal plume hypothesis is that a renewal mechanism is
straightforwardly provided if the plume originates from a thermal boundary layer. It
may be that the effects of water on melting and on plume buoyancy are significant,
but it is far from clear that water alone could provide a sufficient explanation of
the observations, while heat alone, or heat plus water, provides a straightforward
and quantitatively successful account of the dynamical requirements of a theory of
plumes.
×
8.5 Pursuing implications
This completes the discussion of mantle convection
per se
. With the general picture
established, the rest of the topic will explore implications. Potentially there are
many implications, and it is a prime motivation for this topic to provide enough
understanding that these implications can be pursued more readily. However, I will
focus on two that are of fairly fundamental importance, which I have been long
involved with, and on which I am therefore reasonably well informed. These are
the evolution of mantle convection and the chemistry of the mantle.
These topics are closer to the research frontier than the basic mantle convection
covered so far. They will therefore be covered with more reference to the literature
