Melt depletion

Clinopyroxene shows a range of REE patterns from extremely enriched to very depleted TREE signatures (Figure 22). Noncratonic peridotites are subdivided on the basis of clinopyroxene REE patterns into LREE-depleted (type lA) and LREE-enriched (type IB Menzies, 1983 Figure 17). LREE-enriched type IB pyroxenes are the norm in most suites. LREE-depleted varieties are relatively scarce. Very few clinopyroxenes show simple LREE-depleted REE patterns that can be interpreted solely in terms of melt depletion, i.e., LREE depletion, fiat, unfractionated MREE-HREE patterns (e.g., UM-6 or 2905 Eigure 22). For peridotites that do have LREE-depleted clinopyroxenes, a correlation of HREE with other incompatible trace elements (e.g., yttrium, strontium, zirconium) in xenoliths suites worldwide requires fractional melting to be the principal means of depletion in the mantle (Norman, 2001). 

Figure 37 Frequency distribution plots for Os, Nd, and Sr isotope compositions of cratonic and noncratonic peridotite xenoliths. Upper right plots give the range for ocean island basalts (OIB) and arrows show the direction of isotopic evolution for melt depletion and enrichment events. Data compiled from sources cited in Menzies (1990b), Pearson (1999a,b), and those given in Figure 21 (after Pearson and Nowell, 2002).

The robustness of the Lu-Hf isotope system in some mantle environments is demonstrated by the precise Lu-Hf isochron of 1,413 67 Myr dehned by clinopyroxene separates from the Beni Bousera peridotite massif (Pearson and Nowell, 2003). This age probably dates the time of melt extraction from these rocks and is considerably more precise than the Sm-Nd isochron or the scattered Re-Os isotope systematics of these rocks. This indicates the potential power of this system in dating mantle rocks. The initial results from the Lu-Hf isotope system indicate that of the incompatible element isotope systems, it is the more robust to metasomatic effects, with signatures frequently recording the time-integrated response to melt depletion. 

Ducea M., Sen G., Eiler J., and Fimbres J. (2002) Melt depletion and subsequent metasomatism in the shallow mantle beneath Koolau volcano (Hawaii). Geochem. Geo-phys. Geosys. 3 10.1029/200IGCOOO184. 

Handler M. R. and Bennett V. C. (1999) Behaviour of platinum-group elements in the subcontinental mantle of eastern Australia during variable metasomatism and melt depletion. Geochim. Cosmochim. Acta 63, 3597-3618. 

Hanghoj K., Kelemen P. B., Bernstein S., Blustztajn J., and Frei R. (2(X)1) Osmium isotopes in the Wiedemann Fjord mantle xenoliths a unique record of cratonic mantle formation by melt depletion in the Archaean. Geochem. Geophys. Geosys. 2 (20010109) 2000GC000085. 

The effects of post melt-depletion interaction with fluid or melt components in the lithospheric mantle has been extensively documented (e.g. Menzies Hawkesworth 1987, and references therein Harte et al. 1993 Pearson 1999 ) and it is widely accepted that these phenomena dominate the minor element geochemistry of cratonic peridotites. Extensive studies of the effect of metasomatism on the major element chemistry of lithospheric peridotites have also been made (Boyd Mertzman 1987 Keleman et al. 1992, 1998 Walter 1999). To date, most of the discussion has centred around the apparent excess of orthopyroxene, especially in Kaapvaal peridotites. However, major and trace element studies show that it is likely that the abundances of garnet and clinopyroxene are also grossly affected (Burgess Harte 1999 Shimizu 1999). 

Fig. 7. Platinum group element fractions (Pd/Ir)n v. bulk-rock AI2O3 for selected cratonic and circum-cratonic peridotites. Also plotted are the two melt-depletion curves from Fig. 3. It should be noted that some peridotites from Table 1 plot close to the theoretical melting trends (FRB1181 40-21) or on an extension of the trends, at very low Pd/Ir, following complete removal of sulphide. These samples have Tro and Tma ages in reasonably close agreement. Other samples are dispersed away from the trends (see text for details).

The Vitim peridotite with the least disturbed extended-PGE pattern in Fig. 8 sample 314-5, also has similar Trd and Tma ages (1.8 and 2.0 Ga, respectively). The Trd age of 1.8 Ga is the oldest of this suite of peridotites and is similar to the Sm-Nd clinopyroxene model ages in these rocks (Ionov Jagoutz 1989). In contrast to cratonic rocks, the Sm-Nd isotope systema-tics of most Vitim peridotites are characteristic of melt depletion, i.e. clinopyroxenes have Sm/Nd > chondrite and mineral REE patterns do not show the complexity of enriched cratonic rocks (Ionov Jagoutz 1989 Ionov et al. 1993). Hence, in contrast to cratonic rocks (Pearson 19996), the Sm-Nd model ages of Vitim rocks may reflect the partial melting event that produced them. This possibility is supported by the coincidence of Sm-Nd and Re-Os model ages... 

The overabundance of garnet and clinopyr-oxene in most Kaapvaal peridotites (Fig. 1) indicates that these phases probably crystallized some time after melt depletion, during melt-rock (Kele-man et al. 1998) or hydrous fluid-rock reaction (as proposed for some Kaapvaal clinopyroxenes (Boyd Mertzman 1987)). The post-depletion addition of substantial garnet and clinopyroxene would explain the discrepancies in melt models for Kaapvaal peridotites based on comparison of Fe-Mg systematics v. Al-Mg systematics (Walter 1999). 

---