High-pressure melting experiments on garnet clinopyroxenite and the alkalic to tholeiitic transition in ocean-island basalts

Abstract

It has been suggested, on the basis of recent high-pressure melting experiments, that high-Mg garnet clinopyroxenite is the most important lithology controlling the major-element budget of ocean-island lavas [Geology 31 (2003) 481–484]. To clarify these claims and further our understanding of the petrogenesis of ocean-island basaltic (OIB) lavas, we present the results of a high-pressure (2.0–2.5 GPa) melting study on a high-Mg garnet-clinopyroxenite mantle nodule (77SL-582) from Hawaii. Major-element compositions of the partial melts as a function of pressure ( P), temperature ( T), and degree of melting ( F), mineral chemistry of the coexisting crystalline phases, and the solidus/liquidus brackets of this particular garnet clinopyroxenite are reported. The solidus of 77SL-582, which resembles a tholeiitic picrite in terms of its bulk composition, is bracketed at 1295±15 and 1335±15 °C at 2.0 and 2.5 GPa, respectively, which is ∼60–70 °C lower than the solidus of mantle lherzolite at identical pressures [Geochem. Geophys. Geosystems 1 (2000) 2000GC000070]. The solidus of 77SL-582 is also lower by ∼30–40 °C than reported for a slightly alkalic, high-Mg garnet clinopyroxenite [Geology 31 (2003) 481]. At both pressures, the moderate degree (∼18–20%) partial melts of 77SL-582 are strongly alkalic with ∼8–12 wt.% nepheline in the norm. Even at a degree of melting as high as ∼60%, moderately alkalic basaltic liquids are produced. With further melting, however, partial melts become hypersthene-normative. In the CaO–MgO–Al 2O 3–SiO 2 (CMAS) system, the eclogite surface is restricted to the tholeiitic part of the basalt tetrahedron [D.C. Presnall, Effect of pressure on fractional crystallization of basaltic magma, in: Y. Fei, C. Bertka, and B. Mysen (eds.) Mantle Petrology: Field observations and High Pressure Experimentation: A Tribute to Francis (Joe) R. Boyd, The Geochemical Society Special Publication no. 6, 1999, pp. 209–224.]. A comparison with high-pressure melting experiments in the CMAS system at 2.0–3.0 GPa indicates that the alkalic to tholeiitic transition in our experiments can be explained by a rapid expansion of the eclogite “surface” from the tholeiitic part to well into the alkalic part of the basalt tetrahedron in natural systems. Importantly, the partial melting trends of 77SL-582 are transverse to those seen in ocean-island basalts. In addition, although the alkalic to tholeiitic basalt transition observed in ocean-island basalts is well reproduced in our experiments, there is very little overlap between the partial melts of 77SL-582 and ocean-island basalts. It appears that most of the major-element systematics of the ocean-island basalts considered here can be explained by combined contributions of melts from anhydrous and carbonated lherzolite [Geophys. Res. Lett. 24 (1997) 2837; G.H. Gudfinnsson, D.C. Presnall, Continuous gradations among primary kimberlitic, carbonatitic, melilititic, and komatiitic melts in equilibrium with garnet lherzolite at 3–8 GPa, Proc. 8th Int. Kimber. Conf. Extended abstracts.] at high pressures with possible contributions from melts from garnet clinopyroxenite of bulk composition similar to 77SL-582.