Geology Reference
In-Depth Information
However, in the class of models pioneered by Christensen and Hofmann [122],
the heterogeneities are removed by remelting, rather than by being stirred down
to a scale at which the material or melt is homogenised. The empirical success of
such models that has been summarised here indicates that this is the main way in
which heterogeneity is removed in the mantle. The apparent lead age is then to be
interpreted as a mean residence time, not a homogenisation time. The residence
time is the time interval between successive passages through a mid-ocean ridge
melting zone.
Stirring and homogenisation are strongly sensitive to such factors as the time
dependence or dimensionality of the flow [198, 224], and this has motivated much
of the modelling of this process. However, removal by remelting is evidently much
less sensitive to such factors because it depends just on material being carried into or
settling into the two sampling zones, and not on the detailed geometry or topology
of the heterogeneities. This conclusion is supported by the consistency of results
from models with various Rayleigh numbers, various viscosity structures, two-
dimensional or three-dimensional, flat or spherical, and the presence or absence of
plates, all of which would strongly affect the homogenisation process.
This implies that the results obtained here are fairly robust relative to the various
approximations in the models, and this improves our confidence in them.
10.8 Resolving the noble gas enigma
The noble gas observations have been particularly difficult to interpret, for two
main reasons. First, some, but not all, OIBs have unradiogenic helium isotopic
compositions (meaning high
3
He/
4
He or low
4
He/
3
He; Figures 10.4 and 10.16)
compared with MORBs. This has been taken to mean that the source is enriched
in 'primitive' helium - in other words, that it has a higher abundance of the non-
radiogenic
3
He. However, lead isotopes in particular (Figure 10.3) make it clear
that these samples have been previously processed in melting episodes. The helium
would be expected to degas during melting, and the 'primitive' helium to be lost.
Second, estimates of the global budget of
40
Ar, which is the daughter of radioactive
40
K, indicate that about half of it is in the atmosphere, not much is in the continental
crust, and so the balance, about half, should be in the mantle [225]. However, the
MORB source has usually been inferred to have been strongly degassed and to
have a low abundance, insufficient to balance the global budget.
The unradiogenic OIB helium has been a problem for all models, geophysical
or geochemical. The argon mass balance has been dealt with by geochemists by
assuming a large 'undegassed' reservoir somewhere deep in the mantle, initially
identified with the allegedly primitive lower mantle [226]. This would accommo-
date all the argon (and other elements) required by awkward mass balances, but it
