Evolution of Hawaiian Volcano Magmatic Plumbing System and Implications for Melt/Edifice and Melt/Lithosphere Interaction: Constraints from Hualālai Xenoliths

Abstract

Abstract The evolution of Hawaiian magmatic storage and transport systems in response to variations in magma supply over the course of volcano lifespan can have a significant influence on the type and amount of wallrock material that is assimilated by ponded melts prior to eruption. Understanding this plumbing evolution is therefore critical for evaluating the extent to which such melt/wallrock interaction affects the geochemical signals of Hawaiian basalts. We have examined mineral major and trace element and Sr-Nd-Pb-Hf-Os-O isotope variations in a suite of cumulate and lower Pacific crust xenoliths from the Ka‘ūpūlehu flow, Hualālai Volcano, Hawai‘i in order to constrain the depths of magma storage during Hualālai shield- and post-shield-stage volcanism and the effects of edifice and Pacific crust assimilation. Xenoliths range from 1- and 2-pyroxene gabbros to dunites. Pressures of equilibration for gabbroic and pyroxenitic xenoliths, calculated using two-pyroxene and clinopyroxene-only thermobarometry, suggest that most xenoliths, including both shield- and post-shield-stage cumulates, formed within the Pacific lower crust, at pressures >0.24 GPa. However, two gabbros record lower equilibration pressures (<0.2 GPa) and may have formed within the volcanic edifice. Dunite xenoliths also appear to have formed at shallower depths than most gabbro and pyroxenite xenoliths, inconsistent with a single liquid line of descent. These results indicate that, although shallow (intra-edifice) magma chambers are active during Hawaiian shield-stage volcanism, some magmas also pond and fractionate within or near the base of the Pacific crust during the shield stage. Mass and energy constrained geochemical modeling suggests that ponded melts are likely to assimilate significant quantities of wallrock material, with the mass ratio of assimilated material to crystals fractionated approaching one, regardless of depth of ponding. Elevated 187Os/188Os in some evolved post-shield-derived xenoliths are consistent with assimilation of lower Pacific crust, and low δ18O in xenoliths recording shallow equilibration pressures are consistent with edifice assimilation. However, the effects of assimilation on other radiogenic isotopes appear to be limited in most xenoliths and, by inference, in erupted basalts. Melt–wallrock reaction also appears to have modified the composition of the local Pacific crust. Although plagioclase from the lower oceanic crust record unradiogenic Sr-isotopes similar to mid-ocean ridge basalt (MORB), pyroxene Sr-Nd-Hf and whole-rock Os-isotopes have been variably affected by interaction with Hawaiian melts, resulting in a hybrid isotopic composition intermediate between MORB and Hawaiian shield-stage basalts. These hybrid isotopic compositions are qualitatively similar to Hawaiian rejuvenation-stage basalts. Similar hybridization is likely to have altered the isotopic composition of the Pacific lithospheric mantle. Therefore, Pb-isotope differences between MORB and rejuvenation-stage Hawaiian melts do not preclude melt generation within the Pacific lithosphere or asthenosphere. The isotopic signatures of rejuvenation-stage basalts may represent a unique depleted component within the Hawaiian plume, as suggested by previous studies, but requires additional investigation in light of these results.