Abhatoo - Centre Nationale de Documentation

A fundamental challenge in modern mantle geochemistry is to link geochemical data with geological and geophysical observations. Earlier efforts have taken the approach that large scale regions in the mantle, such as the “convective” (i.e. well-stirred, homogeneous) upper mantle, sub-continental lithosphere, and the lower mantle are distinct geochemical reservoirs. Here we present evidence for a ubiquitous distribution of small-to-moderate scale heterogeneity in the upper mantle, the ‘Statistical Upper Mantle Assemblage’, which forms as the result of long-term plate tectonic recycling of sedimentary and crustal components. This model differs from previous models of upper mantle heterogeneity, including the ‘plum pudding’ and the ‘marble cake’ models, in at least two critical ways. The Statistical Upper Mantle Assemblage avoids the necessity of a convectively homogenized and depleted MORB mantle reservoir, which is a fundamental concept in all previous models. Instead, we explain the isotopic compositions of mid-ocean ridge and oceanic island basalts as the predicted outcome of statistical sampling of a distribution of upper mantle heterogeneity consisting of distinct components. Furthermore, the Statistical Upper Mantle Assemblage model and the methods outlined in this work provide a strong link between geochemical data and potential geophysical observables, such as the degree of partial melting, the characteristic length scales of the melt system and chemical heterogeneity, bathymetry, seismic scattering and anisotropy. We suggest that statistical distributions of lithologic components replace the concept of distinct, isolated reservoirs.