Geology Reference
In-Depth Information
separate episode [239]. Bombardment persisted until about 3.8 Ga, by which time
about one-third of the Earth's
40
Ar had been generated. If mantle degassing was
rapid during the first 0.5 Gyr or so, then it is conceivable that the order of 10% of
the Earth's
40
Ar was removed from the Earth entirely [240].
Thus the
40
Ar constraint is not as stringent as has been claimed. There are
uncertainties in the amount of potassium in the continental crust and in the bulk
silicate Earth, and there may have been some early loss of
40
Ar from the atmosphere.
The present mantle model may account for the total
40
Ar budget of the Earth when
these uncertainties are allowed for.
10.8.7 Alternative interpretations
The presence of
3
He in some samples was never a reason in itself for inferring a
primitive source. Certainly
3
He is not produced in the Earth, so any that remains
is 'primitive', but it is hardly surprising that some small amount is still trickling
out of the Earth. The real question is how the present amount compares with
plausible initial amounts, and with the amount of
4
He produced in the meantime,
questions addressed implicitly in the interpretation just presented. The jump to
presuming an 'undegassed' mantle reservoir, meaning one that has retained all
radiogenic gases produced over 4.5 Gyr, was evidently conditioned by the pre-
sumption of a 'primitive' lower mantle, originally proposed to explain a handful of
dubious Nd data. That presumption has failed to survive critical scrutiny, as we have
seen.
A recent proposal by Tolstikhin and Hofmann [241] is not so readily dismissed.
They proposed that D
was formed very early in Earth history from a foundered
mixture of late-accreting material and early basaltic crust. The accreting material
is presumed to carry high concentrations of solar noble gases and excess iron.
The excess iron would increase the density and thus tend to stabilise the layer and
allow it to survive to the present. This model deals better with the main constraints,
though it requires rather detailed specification of the initial state of D
and may
be difficult to test. Nevertheless, it has more merit than many predecessors and
deserves to be critically evaluated.
Albarede [235] suggested that unradiogenic helium comes from sources with
low U, rather than high He, and that such a source would be produced if He was
less incompatible than U during melting. However, the source would then be a
melt residue that would be refractory and have very low concentrations of both
elements, so the helium would not be readily evident in subsequent melting events.
The residue would also tend to be less dense than average mantle, so it is not obvious
why it would show up preferentially in mantle plumes. Some other hypotheses that
have been proposed at various times are discussed by Davies [227].
