Mantle plumes sample heterogeneous mixtures of oxidized and reduced lithologies

Abstract

Constraining the heterogeneity of oxygen fugacity (fO2) in the upper mantle is important for understanding the role of the deep Earth in determining the redox state of the atmosphere. Measuring the fO2 of primitive, mantle-derived basalts is a means with which to constrain mantle fO2 heterogeneity. Trace element systematics of 721 primitive olivine crystals from fifteen Icelandic picrites are reported, and these data are used with previously published bulk rock compositions to constrain their magmatic fO2. These fO2 values were compared with previously reported ocean island basalt (OIB) fO2 data calculated using identical methods, showing that fO2 is negatively correlated with parental magma MgO and positively correlated with bulk rock La/Yb, trends which were not evident in previous, smaller, global ocean island basalt datasets. These correlations cannot reasonably be the result of fractional crystallization, differential volcanic degassing, or differential partial melting of a typical homogenous mantle peridotite. Instead, differential partial melting of a heterogeneous source which preferentially sampled oxidized fusible lithologies at low melt fractions and diluted such signatures with depleted peridotite melts at high melt fractions likely caused the observed fO2-MgO-isotopic trends. These conclusions further reinforce that the mantle should not be modeled as a homogenous peridotite, but instead as a heterogeneous, non-equilibrated mixture of enriched “blebs” in a matrix of depleted peridotite, and that degree of partial melting is an important control on what portions of the mantle are detectable in igneous rocks.