Sulfur isotope characterization of primordial and recycled sources feeding the Samoan mantle plume

International audienceThe Samoan islands are the type locality for the Enriched Mantle (II) reservoir that is thought to be produced from the subduction and recycling of marine sediment from upper continental crust. In addition to hosting extreme radiogenic isotope compositions from the EM II reservoir, Samoa also exhibits contributions from other mantle reservoirs in a dilute form including the EM (I) (recycled continental material), HIMU (recycled oceanic crust), and DMM (depleted upper mantle) mantle reservoirs. The plume system feeding the Samoan islands sits above a seismically imaged Large Low Shear Velocity Province (LLSVP) and an Ultra-Low Velocity Zone (ULVZ) that is thought to contribute, in addition to recycled components, the recently discovered early-formed (primordial) components with negative µ182W and high 3He/4He. Recent work measuring sulfur isotopes in ocean island basalts has established that recycled oceanic and continental crust host unique S-isotope compositions that can be identified at various hotspot localities. Here we document previously unknown relationships between Δ33S and radiogenic tungsten, helium and lead isotopes from 7 Samoan basalts (from the islands of Ofu, Vailulu’u and Malumalu) that suggest mixing between several endmembers. One, a HIMU influence that has slightpositive Δ33S and positive δ34S; another, related to EM II that has near zero Δ33S and positive δ34S ; a third, which is primordial with negative µ182W, high 3He/4He, that has Δ33S=0 and negative δ34S. From this, we conclude that the indistinguishable Δ33S of the primordial endmember from that of the convective mantle indicates that sulfur isotopes were homogenized early in Earth’s history. The Vailulu’u sample with HIMU characteristics, carries a small but resolvable Δ33S that allows, but does not require mass-independent Archean Δ33S to shift the Δ33S. The observed correlations involving Δ33S supports 38 arguments linking Pb, He, and W geochemistry to a deep mantle process and places constraints on questions related to the sources of mantle geochemical heterogeneity