Geochemical Evolution Of The Mantle

Woods Hole Oceanographic Institution, Woods Hole MA

Investigators

Abstract

Hart EAR-0125917 This grant has as its overarching goals the understanding of the placement, genesis, evolution and lithologic nature of the several known heterogeneous mantle domains, and the dynamics of melting and melt extraction from these mantle domains. Several paradigm "warps" have been in progress during the period of our last renewal. One is the increasing evidence that mesoscale mantle heterogeneities (veins, pods, layers) may be important in creating some of the isotopic heterogeneity observed in mantle-derived melts. This in turn demands a better understanding of melting and melt transport processes. The second is the concept that the deepest mantle may be convectively isolated from the overlying mantle (with a boundary at circa 1700 km). This in turn demands a better understanding of the chemical and lithologic make-up of mantle plumes, as they may provide our most direct sampling of the deep mantle. Several of the isotopically-defined mantle domains are thought to be a result of lithospheric recycling, and yet we have little understanding of the parentage or lithologic expression of this material, or of its melting dynamics. We will explore two avenues as a means to confront this problem. The first is an in-depth study of the Samoa hotspot (the most extreme EM2 mantle plume). The high 87Sr/86Sr (~ 0.7089) of the Samoan plume, and its distinctive chemistry, is conventionally ascribed to recycling of ancient lithosphere containing a terrigenous sediment component. However, no model has as yet satisfactorily accounted for many of the puzzling geochemical characteristics of EM2 hotspots (e.g. the coexistence of enriched 87Sr/86Sr, heavy d18O, and high 3He/4He). We propose to undertake analytical studies of volcanics from the older seamounts west of Samoa, of shield lavas from western Upolu and Savai'i (putatively the oldest subaerial volcano in Samoa), and of basalts from two younger seamounts east of Samoa. Sample suites from all of these locations are in hand. This work will include 40Ar/39Ar dating of these volcanoes, to test the age-progression hypothesis, and trace element and isotopic work to elucidate the pedigree of the EM2 component in Samoa. The second avenue is a study of the isotopic composition and trace element signatures of melt inclusions in phenocrysts from EM1 and EM2 basalts (using in-situ ionprobe and laser-ablation ICP/MS techniques). In general, the geochemical diversity captured in melt inclusions far exceeds that observed in the erupted host lavas. The nature of these melt inclusions, which likely represent pre-aggregated melts, will shed significant light on the melting lithologies and melt transport networks involved in EM hotspot volcanism. We should be able to discriminate between large-scale and small-scale mantle heterogeneities, and between mafic and ultramafic lithologies; this should allow a firm test of the recycled lithosphere model for EM2 plumes.