Olivine megacryst chemistry, Monastery kimberlite: Constraints on the mineralogy of the HIMU mantle reservoir in southern Africa

Abstract

Olivine phenocryst chemistry is a useful tracer of mantle source lithology as olivine is typically the first mineral to crystallize in a range of mafic to alkaline magma types and contains geochemical information about the primary parent magma composition. It is particularly useful in discriminating between pyroxenite (i.e.,recycled crustal component) and peridotite mantle source lithologies. Radiogenic isotope studies of HIMU basalts have shown the important role of a recycled crustal component in the source. However, olivine chemistry of HIMU basalts suggests a dominantly peridotitic mantle source with a subduction-derived metasomatic carbonate component. To further investigate how olivine chemistry can be used to understand the source of HIMU magmas, I present major and trace element data for olivine megacrysts related to the 89 Ma Cr-poor megacryst suite from the Monastery kimberlite (South Africa), which have previously been interpreted to crystallize at high P-T conditions of ~1400 °C and ~5 GPa from a magma sourced from a HIMU reservoir. Olivine megacrysts have high-Ni concentrations at a given Fo (forsterite) content, overlapping ocean island basalts (OIB) interpreted to have formed from pyroxenite-dominated sources but are distinct from typical olivine in HIMU basalts. However, they have low 100*Mn/Fe (0.8–1.1) and no correlation is observed between Ni and trace elements indicative of recycled components such as Co, Li, or Zn. The olivine megacryst chemistry is similar to that of aillikite olivine interpreted to be controlled by phlogopite in the source rather than pyroxenite. Comparison with olivine chemistry from orangeites/Group II kimberlites (sourced from phlogopite-rich mantle lithologies) supports a phlogopite controlon low 100*Mn/Fe in olivine. Further comparison with olivine phenocryst chemistry of HIMU melilitites (76–58 Ma) in southern Africa suggests a mineralogically heterogeneous HIMU reservoir formed by metasomatic modification of the lithospheric mantle in southern Africa. Thus, olivine megacryst chemistry supports interpretations for the recycled component of the HIMU reservoir as mineralogically complex metasomatic lithologies formed by the infiltration of subduction-derived melts into the base of the lithospheric mantle. In addition to carbonate-richHIMU mantle lithologies in southern Africa and worldwide, Monastery kimberlite olivine megacrysts and olivine in melilitites suggest that a phlogopite-richHIMU lithology is present in southern Africa. An important question arising from this study is in reconciling the homogenous isotopic ratios of HIMU basalts worldwide with an apparent heterogeneous lithological source indicated by olivine chemistry.