Late veneer

Water and other volatiles could have been supplied to Earth by comets and asteroids as part of the late veneer. The arguments for and against this hypothesis have recently been reviewed by Drake (2005). The D/H ratio measured in three comets to date is 2 x higher than on Earth, suggesting that comets could not have supplied more than 50% of Earth s water (Robert 2001). However, these comets may not be representative of objects colliding with the early Earth. If the Ar/H20 ratio measured in comet Hale-Bopp is typical, comets would have delivered 2 x 104 times more Ar than is presently found in Earth s atmosphere if they were the main source of Earth s water (Swindle Kring 2001). Consideration of the abundances of noble metals and noble gases led Dauphas Marty (2002) to estimate that comets contributed <1% of the Earth s water. It is unlikely that carbonaceous chondrites supplied most of the late veneer since these objects have different Os isotope ratios than Earth s mantle,... 

Figure 1 The estimated composition of the silicate portion of the Earth as a function of condensation temperature normalized to Cl values in Anders and Grevesse (1989). Open circles lithophile elements shaded squares chalcophile elements shaded triangles moderately siderophile elements solid diamonds highly siderophile elements. The spread in concentration for a given temperature is thought to be due to core formation. The highly siderophile element abundances may reflect a volatile depleted late veneer. Condensation temperatures are from Newsom (1995).

Thus core-mantle equilibration can be excluded as the source of the HSEs in the Earth s mantle. It is more likely that a late accretionary component has delivered the HSEs to the Earth s mantle, either as single Moon-sized body which impacted the Earth after the end of core formation or several late arriving planetesimals. The impac-tors must have been free of metallic iron, or the metallic iron of the projectiles must have been oxidized after the collision(s) to prevent the formation of liquid metal or sulfide that would extract HSEs into the core of the Earth. The relative abundances of the HSEs in the Earth s mantle are thus the same as in the accretionary component, but may be different from those in the bulk Earth. The late addition of PGE with chondritic matter is often designated as the late veneer hypothesis (Kimura et al., 1974 Chou, 1978 Jagoutz et al., 1979 Morgan et al., 1981 O Neill, 1991). This model requires that the mantle was free of PGE before the late bombardment established the present level of HSEs in the Earth s mantle. 

The late veneer hypothesis has gained additional support from the analyses of the osmium isotopic composition of mantle rocks. Meisel et al. (1996) determined the Os/ Os ratios of a suite of mantle xenoliths. Since rhenium is more incompatible during mantle partial melting than osmium, the Re/Os ratio in the mantle residue is lower and in the melt higher than the PM ratio. By extrapolating observed trends of Os/ Os versus AI2O3 and lutetium, two proxies for rhenium, Meisel et al. (1996) determined a Os/ Os ratio of 0.1296 0.0008 for the primitive mantle. This ratio is 2.7% above that of carbonaceous... 

The small amount of the late veneer (<1% chondritic material) would not have had a measurable effect on the abundances of other elements besides HSEs, except for some chalco-phile elements, most importantly sulfur, selenium and tellurium (Figure 15(c)). The amount of sulfur presently in the Earth s mantle (200 ppm. Table 4) corresponds to only 0.37% of a nominal CI-component, while the iridium content suggests a CTcomponent of 0.67%. O Neill (1991) has, therefore, suggested that the late veneer was compositionally similar to H-chondrites which contain only 2% S (Wasson and Kallemeyn, 1988). Because H-chondrites have higher iridium (780 ppb) than Cl-chondrites, the required... 

H-chondrite fraction would only be 0.41% based on iridium. This would correspond to 82 ppm S delivered by the late veneer to the mantle. In this case the Earth s mantle should have combined —120 ppm S before the advent of the late veneer. If core formation in the Earth (or in differentiated planetesimals that accreted to form the Earth) occurred while the silicate portion was molten or partially molten, some sulfur must have been retained in this melt (O NeiU, 1991). 

The oldest samples that provide direct constraints on the osmium isotopic composition of the upper mantle are rare spinel peridotites contained within the Early Archean Itsaq gneiss complex of southwest Greenland that are interpreted to be —3.81 Ga abyssal peridotites (Friend et al., 2002). The measured and initial compositions determined from low-Re/Os spinel and olivine mineral separates from these peridotites are the most primitive, in the sense of closest to solar-system initial compositions, Os isotopic compositions yet obtained on any terrestrial material (Bennett et al., 2002). This shows that at least some, if not all, of the Early Archean upper mantle was characterized by chondritic Os/ Os isotopic compositions. Osmium isotopic constraints from this time period (—3.8-3.9 Ga) are of particular interest as they provide a rough constraint on the timing of the addition of the late veneer of... 

The amount of nitrogen in the upper mantle is therefore 1.6 X 10 ° g, which is only 3% that in the crust and atmosphere. The MORE flux is equivalent to a flux of 5.0 X 10 °molyr or 9% of the surface N2 over 4.5 Ga. Note that Javoy (1998) argued that nitrogen is relatively compatible, with an upper-mantle concentration of up to 40 ppm (Cartigny et ai, 2001). However, this is from mass balance calculations based upon model assumptions that volatiles in the upper mantle and on the surface are a mixture of enstatite chondrites and a late veneer of Cl chondrites, and such a model has not been widely adopted. 

The accretion of a late veneer. Whether or not a late veneer was added to the Earth towards the end of accretion is not clear. The principal evidence comes from the elevated siderophile element chemistry of the mantle. If there was a late veneer, it had to happen after core formation. At present the evidence from the trace element chemistry is ambiguous, because these data can also be explained by the formation of the core at high pressures and temperatures in a magma ocean, or by continuous core formation with decreasing metal input. Currently, the best evidence for a late veneer comes from Os-isotope evidence, where there is a clear mismatch between the composition of the PUM and chondrite. However, even this is uncertain, as is discussed in the next chapter (Chapter 3, Section 3.2.3.4), for... 

Dauphas, N. and Marty, B., 2002. Inference on the nature and mass of the Earth s late veneer from noble gases and metals, f. Geophys. Res., 107, doi 10.1029/2001JE001617. 

---