ERI Summer 2017 Fellowship Report

Introduction Ocean island basalts (OIB) are thought to sample the Earth’s deep mantle, whereby thermochemical plumes entrain lower mantle material and bring it to the surface where it erupts. By measuring the isotopic compositions of global OIB—which record the isotopic compositions of OIB mantle sources—geochemists can “map” the isotopic heterogeneity of the Earth’s deep mantle (e.g., Gast et al., 1964; Hart et al., 1973; Zindler and Hart, 1986). OIB can be grouped into distinct endmembers defined by their Sr, Nd, and Pb isotopic compositions: the DMM (depleted mid-ocean ridge basalt [MORB] mantle), HIMU (high μ = U/Pb), EM-1 (Enriched Mantle 1), and EM-2 (Enriched Mantle 2) end-members are often used to categorize OIB and their respective mantle sources (Zindler & Hart, 1986; Figure 1). Although it is clear that significant isotopic differences exist in the mantle domains sampled by global OIB, the origins and geometries of such isotopic “reservoirs” in the Earth’s mantle remain active areas of inquiry. Many prior geochemical studies of mantle-derived lavas have implemented whole-rock isotopic measurements that necessarily homogenize any heterogeneity hosted between different mineral phases or within individual minerals. Although whole-rock analyses are useful in identifying large-scale isotopic heterogeneities preserved in the mantle (e.g., Hofmann, 1997; Stracke, 2012; White, 2015), such analyses erase potentially informative heterogeneity hosted within a single lava (Davidson et al., 2007). Recent work has identified Sr/Sr heterogeneity between olivine-hosted melt inclusions and their host OIB whole rocks, suggesting that multiple, isotopically distinct melt sources can contribute to a single lava (e.g., Jackson & Hart, 2006; Sobolev et al., 2011; Reinhard et al., 2016). Similar work has identified disequilibrium between whole-rock lavas and magmatic cpx in Samoan OIB (Jackson et al., 2009). However, the origins of such isotopic heterogeneity within a single erupted lava remain enigmatic. Plagioclase is well suited to evaluating the origin of isotopic heterogeneity in OIB because plagioclase often contains zones of distinct isotopic and chemical compositions. Such zones record a crystal “stratigraphy” that can provide information about the melt(s) contributing to a single magma chamber—and a single lava—over time. Thus, plagioclase represents an ideal target for geochemical analyses to address the processes that lead to the isotopic disequilibrium that has been previously observed in OIB.