Constraints on Lithospheric Mantle Evolution From
Re-Os Isotope Studies of Xenoliths

D. G. Pearson Dept. of Geological Sciences, University of Durham, Durham, DH1 3LE, UK

d.g.pearson@durham.ac.uk

Cratonic Xenoliths

Walker et al. (1989) first applied the Re-Os system to a suite of peridotite xenoliths from the Kaapvaal Craton and found very unradiogenic 187Os/188Os values requiring long term evolution in low Re/Os environment. These authors recognised that Re addition has occurred in many kimberlite-borne xenoliths and developed the concept of Re depletion ages (TRD) which assume initial formation with Re/Os = 0 and are thus minimum age estimates.

Peridotite xenoliths from 4 different cratons have been analysed so far (Walker et al., 1989; Pearson et al., 1995a,b,c; Carlson and Irving, 1994). These sample suites are characterised by a range of gOsT values (% difference from Bulk Earth) ranging from slightly above 0 to very unradiogenic values (as low as -17) that require ancient, Mid Archaean minimum formation ages. The lowest gOsT values for peridotites from the Kaapvaal (Walker et al., 1989; Pearson et al., 1995a), Siberian (Pearson et al., 1995b) and Wyoming (Carlson and Irving, 1994) cratons are -17 to -15. These values translate into TRD ages of 3.5, 3.3 and 2.9 Ga for the respective cratons and overlap the periods of major crust building on each craton. The large spread in gOsT values for each sample suite is thought at present to represent the time-integrated effects of Re and Os enrichment from the Archaean to present day.

Three peridotites from the Kirkland Lake kimberlite (Pearson et al., 1995c), erupted through Late Archaean crust have gOsT values of -7 to -13. Although the sample set is small, the oldest TRD age of 2.6 Ga matches well with the age of major crust building and greenstone formation in the Abitibi Belt.

Pyroxenite xenoliths from cratonic regions appear to record ancient melt infiltration events within the lithosphere (Carlson and Irving, 1994). The significance of eclogites is still being debated but a 2.9 ± 0.4 Ga isochron for a suite of Siberian eclogites (Pearson et al., 1995d) with model ages up to 3.5 Ga confirms the ancient, Mid Archaean age for the Siberian CLM. If a subducted protolith is accepted imply deep Archaean subduction. Eclogites from S. Africa show much more complex, multi-stage Os systematics but Re and Os abundances are consistent with genetic groupings (Pearson et al., 1992).

Xenoliths erupted through Proterozoic crust

A suite of 19 Namibian peridotites erupted through Proterozoic basement, off craton in southern Africa (Pearson et al., 1994) have gOsT values as low as -11 with a mean of -7 compared to a mean of -10 for peridotites from beneath the Kaapvaal craton. TRD ages range up to 2.1 Ga, coincident with the oldest basement in southern Namibia. The coupling of major crust building and lithospheric mantle ages on and off craton in southern Africa suggests a genetic relationship and very long term stability of deep CLM beneath and around cratons in this region.

Xenoliths from Ocean islands and young continental rifts

2 peridotite xenoliths from Pacific hotspots (Hauri et al., 1993) have gOsT values of +1, i.e., close to Bulk Earth, consistent with an origin as fragments of oceanic lithosphere, rather than from the hotspot source. 2 spinel lherzolites from Kilbourne Hole (Luck and Allegre, 1991; Hart and Ravizza, 1993) have gOsT values close to 0. The xenoliths could be interpreted as recently accreted asthenosphere/MORB source mantle but older ages, similar to the 1.3 Ga regional crust are not precluded.

Future directions

Os signatures are not as immune from metasomatism
as first thought and it appears that suites of 15 or more
peridotites are required to be confident of sampling less disturbed samples. Much work remains to be done: a) evaluating the significance of model age ranges; b) establishing the location of Re and Os in mantle samples; and
c) sulphides behaviour during xenolith eruption. Further studies are required to determine crust-mantle ages in different tectonic settings in order to guide our thinking of how crust is generated and recycled.

References

Carlson, R.W. & Irving, A.J., Earth Planet. Sci. Lett. 126, 457-472, (1994).

Hart, S.R. & Ravizza, G., EOS (Abstr.) 74, No. 14, 120 (1993).

Hauri, E.H. et al., Nature 365, 221-227 (1993).

Luck, J.M. & Allegre, C.J. Earth Planet. Sci. Lett. 126, 406-415, (1991).

Pearson, D.G. et al., EOS (Abstr.) 73, No. 14, 376 (1992).

Pearson, D.G. et al., Min. Mag. 58A, 703-704 (1994).

Pearson, D.G., et al., Earth Planet. Sci. Lett. 134, 341-357, (1995a).

Pearson, D.G., et al., Geochim. Cosmochim. Acta 59, 959-977 (1995b).

Pearson, D.G. et al., Extd. Abs. 6th Int. Kimb. Conf. 427-429 (1995c).

Pearson, D.G. et al., Nature 374, 711-713 (1995d).

Walker, R.J. et al., Geochim. Cosmochim. Acta 53, 1583-1595 (1989).