A trace element perspective on the source of ocean island basalts (OIB) and fate of subducted ocean crust (SOC) and mantle lithosphere (SML)

Abstract

We analyze the first-order observations, basic concepts and explicit/implicit assumptions built into the three major hypotheses for the enriched component(s) in the source of ocean island basalts (OIB) in terms of incompatible trace elements: (1) subducted ocean crust (SOC), (2) subducted continental sediments, and (3) mantle metasomatism. SOC is compositionally too depleted (i.e., [La/Sm]N <1) to be the major source material for OIB that are highly incompatible element enriched (e.g., [La/Sm]N »1). We cannot rule out the contribution of continental sediments as an enriched OIB source component; however, except for two known cases that are yet to be further investigated, there is no convincing evidence for any significant sediment contribution to the petrogenesis of global OIB. Continental materials through subduction erosion can certainly contribute to mantle compositional heterogeneity and may contribute towards some enriched component in OIB source regions. Overall, OIB are not only enriched in incompatible elements, but also enriched in the progressively more incompatible elements, with their inferred source material being variably more enriched than the primitive mantle. These observations require that the OIB sources are pre-enriched by lowdegree (Low-F) melt metasomatism. The interface between the growing oceanic lithosphere and the top of the seismic low velocity zone (LVZ) represents a natural peridotite solidus and is the ideal site for major low-F melt induced metasomatism. The ∼ 70 Myr history of oceanic lithosphere growth to its full thickness of ∼ 90 km records the history of mantle metasomatism, resulting in the deep portion of the oceanic lithosphere being an important enriched geochemical reservoir. We argue that ancient subducted metasomatized mantle lithosphere (SML) provides the major source component for OIB. The metasomatic agent is an H2O-CO2-rich silicate melt derived from within the LVZ. Upward migration and concentration of the melt at the lithosphere-LVZ interface (i.e., the lithosphere-asthenosphere boundary or LAB) results in chemical stratification in the LVZ with the deeper portion being more depleted (i.e., DMM), providing the source for MORB. The widespread metasomatized peridotites, pyroxenites and hornblendites from xenolith suites exhumed from the deep lithosphere (both oceanic and continental) and in orogenic peridotite massifs confirm the role of a low-F silicate melt phase as the metasomatic agent. The SOC, if subducted into the lower mantle, will be too dense to return in bulk to the upper mantle source regions of oceanic basalts, and may have contributed to the two large low shear wave velocity provinces (LLSVPs) at the base of the mantle beneath the Pacific and Africa over Earth's history.