Clinopyroxene and Garnet Mantle Cargo in Kimberlites as Probes of Dharwar Craton Architecture and Geotherms, with Implications for Post-1·1 Ga Lithosphere Thinning Events Beneath Southern India

Abstract

AbstractThe Wajrakarur Kimberlite Field (WKF) on the Eastern Dharwar Craton in southern India hosts several occurrences of Mesoproterozoic kimberlites, lamproites and ultramafic lamprophyres, for which mantle-derived xenoliths are rare and only poorly preserved. The general paucity of mantle cargo has hampered the investigation of the nature and evolution of the continental lithospheric mantle (CLM) beneath cratonic southern India. We present a comprehensive study of the major and trace element compositions of clinopyroxene and garnet xenocrysts recovered from heavy mineral concentrates for three c.1·1 Ga old WKF kimberlite pipes (P7, P9, P10), with the goal to improve our understanding of the cratonic mantle architecture and its evolution beneath southern India. The pressure-temperature conditions recorded by peridotitic clinopyroxene xenocrysts, estimated using single-pyroxene thermobarometry, suggest a relatively moderate cratonic mantle geotherm of 40 mW/m2 at 1·1 Ga. Reconstruction of the vertical distribution of clinopyroxene and garnet xenocrysts, combined with some rare mantle xenoliths data, reveals a compositionally layered CLM structure. Two main lithological horizons are identified and denoted as layer A (∼80–145 km depth) and layer B (∼160–190 km depth). Layer A is dominated by depleted lherzolite with subordinate amounts of pyroxenite, whereas layer B comprises mainly refertilised and Ti-metasomatized peridotite. Harzburgite occurs as a minor lithology in both layers. Eclogite stringers occur within the lower portion of layer A and at the bottom of layer B near the lithosphere–asthenosphere boundary at 1·1 Ga. Refertilisation of layer B is marked by garnet compositions with enrichment in Ca, Ti, Fe, Zr and LREE, although Y is depleted compared to garnet in layer A. Garnet trace element systematics such as Zr/Hf and Ti/Eu indicate that both kimberlitic and carbonatitic melts have interacted with and compositionally overprinted layer B. Progressive changes in the REE systematics of garnet grains with depth record an upward percolation of a continuously evolving metasomatic agent. The intervening zone between layers A and B at ∼145–160 km depth is characterized by a general paucity of garnet. This ‘garnet-paucity’ zone and an overlying type II clinopyroxene-bearing zone (∼115–145 km) appear to be rich in hydrous mineral assemblages of the MARID- or PIC kind. The composite horizon between ∼115–160 km depth may represent the product of intensive melt/rock interaction by which former garnet was largely reacted out and new metasomatic phases such as type II clinopyroxene and phlogopite plus amphibole were introduced. By analogy with better-studied cratons, this ‘metasomatic horizon’ may be a petrological manifestation of a former mid-lithospheric discontinuity at 1·1 Ga. Importantly, the depth interval of the present-day lithosphere–asthenosphere boundary beneath Peninsular India as detected in seismic surveys coincides with this heavily overprinted metasomatic horizon, which suggests that post-1·1 Ga delamination of cratonic mantle lithosphere progressed all the way to mid-lithospheric depth. This finding implies that strongly overprinted metasomatic layers, such as the ‘garnet-paucity’ zone beneath the Dharwar craton, present structural zones of weakness that aid lithosphere detachment and foundering in response to plate tectonic stresses.