Answers are grounded in this paper and cite their sources.
Oceanic peridotites preserve records of both recent melting at mid-ocean ridges and ancient melt depletion, providing key insights into the long-term chemical evolution of the Earth’s mantle. Here we report mantle xenoliths from Tahiti and Moorea in the Society Islands, which sampled the ~70 Myr-old Pacific lithosphere. These samples provide a unique opportunity to investigate the compositional structure and evolution of oceanic mantle far from the mid-ocean ridges. Among 40 investigated xenolith samples (3 lherzolites, 11 harzburgites, 20 dunites, 4 wehrlites, and 2 orthopyroxenites), refractory harzburgites yield unradiogenic 187 Os/ 188 Os ratios down to 0.11720, while the other lithologies are more radiogenic (0.12324–0.13945), displaying a bimodal distribution of Os isotopic compositions. The dunites, wehrlites, and orthopyroxenites are interpreted to have formed through extensive interactions between peridotites and infiltrating melts. High temperature (> 1000 °C) harzburgites and lherzolites host abundant CO 2 -rich fluid inclusions, suggesting metasomatic imprints associated with upwelling mantle plume-derived melts or fluids. The unradiogenic Os isotopic compositions of the refractory harzburgites indicate the presence of an ancient, depleted mantle domain that predates the formation of the ~70 Ma Pacific lithosphere. We interpret these refractory harzburgites as fragments of an ancient refractory mantle domain incorporated into the Pacific lithosphere. One possible origin of this mantle domain is transport and emplacement by the Pitcairn–Gambier plume located southeast of the Society hotspot. In this scenario, the refractory mantle domain was emplaced at the base of the young Pacific lithosphere, became incorporated into the lithospheric mantle, and was subsequently sampled as xenoliths during later Society plume volcanism. These observations highlight the role of mantle plumes in modifying and assembling heterogeneous mantle domains within oceanic lithospheric mantle through time.
No takes yet. Share an insight, caveat, or question.
Akizawa et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: