Metasomatism of the Kaapvaal Craton during Cretaceous intraplate magmatism revealed by combined zircon U-Pb isotope and trace element analysis

Abstract

The origins of the mica-amphibole(K-richterite)-rutile-ilmenite-diopside (MARID) and the, presumably related, phlogopite-K-richterite-peridotite (PKP) mantle xenolith suites remain enigmatic. For instance, it is unclear if MARID rocks represent pervasively metasomatised peridotite or veins of magmatic material. In the Kaapvaal Craton, previous studies employing zircon U-Pb dating in MARID and PKP xenoliths demonstrated that ages of zircon are broadly coincident with nearby expressions of Cretaceous kimberlite and orangeite magmatism, implying a genetic link. However, the use of typically small sample sizes (i.e. few analysed grains) in previous studies may mask complexities in the growth history of zircon in mantle lithologies, as key populations may remain undetected. In this paper, we analyse the isotopic (U-Pb) and trace element composition of zircon (n > 40) from three MARID/PKP rocks from the Kimberley region of the Kaapvaal Craton, in order to reveal not just the timing, but the tempo of metasomatism. This protocol takes its inspiration from studies of detrital zircon in clastic sedimentary rocks, in which large numbers of grains are typically dated, so that no important parts of the zircon U-Pb record are missed.

Our data show that zircon growth is broadly coincident with nearby expressions of Cretaceous kimberlite magmatism and orangeite magmatism. While the range of zircon U-Pb ages we detect mirrors the range detected by previous workers, the addition of larger sample sizes means we can detect that zircon growth was episodic. By integrating trace element data with zircon ages, we demonstrate that older zircon, generally has greater geochemical affinity to lamproitic zircon. By contrast, most zircons with ages coincident with peak kimberlite magmatism have trace-element compositions similar to megacrystic zircon - a suite of zircon crystals thought to be connected to early high-pressure proto kimberlitic liquids/fluids.

These data strongly indicate that MARID and PKP rocks form during localized and recurrent melting of enriched lithosphere. The metasomatizing effects of the passage of varying magmas at shallow depth in the lithosphere may also contribute to their complex age and geochemical signatures. The presence of zircon with U-Pb ages that span the entire range of Cretaceous intra-cratonic magmatism nearby to Kimberley, and trace element compositions that derive from both major phases of intracratonic magmatism in a single MARID xenolith, necessitates a more complex origin for the MARID suite than that of a simple magmatic cumulate. This observation argues for a complex origin for zircon in MARID and presumably PKP rocks.