Water and other Volatiles in the Hawaiian Plume: A Sims Study of Melt Inclusions

Abstract

The importance of water in many mantle processes cannot be overestimated. Water has a substantial influence on lowering the melting temperature of the mantle (Kushiro, 1972), and even small amounts of water in olivine (50 ppm) can lower the viscosity of this mineral by almost three orders of magnitude (Hirth and Kohlstedt, 1996). Because of the significant role of water in modulating viscous mantle flow as well as depth and extent of mantle melting, it is imperative to understand the distribution of water in the Earth's interior, the origin of water in various mantle reservoirs, and the cycling of water between these reservoirs. Hawaii is especially important in studies of mantle water, due to the presence of several distinct mantle components identified from geochemical studies of shield-stage basalts. One of these components is recognized to have a substantial lprimordial' volatile signature (Loihi seamount), while another exhibits many isotopic and geochemical features expected from subducted and recycled crustal material (Koolau volcano). As a result, studies of Hawaiian lavas provide a means of accessing two potentially distinct sources of water in the Earth's interior, and may permit a search for water which has survived passage through a palaeo-subduction zone. Using methods developed for microanalysis of the volatiles H20, CO2, S, F and C1 by ion microprobe, we have addressed the origin of water in the subHawaiian mantle by examining the geochemistry of small inclusions of melt trapped within olivine phenocrysts from a number of Hawaiian volcanoes (Loihi, Kilauea, Mauna Loa, Koolau). In addition, we have recently developed the capacity to measure by ion probe the D/H ratios of magmatic water contained within these melt inclusions, thereby providing an important isotopic tracer for various processes (both deep and shallow) involving water. All analyses are made with a primary beam of Cs + ions (1-10 nA) and collection of negative secondary ions. Detection limits are 20-30 ppm H20, 5 ppm CO2, and <5 ppm for S, F and C1. Reproducibility of D/H ratios in glasses with 1% to 0.2% H20 ranges from 21 to 51 (1~) in a 20-40 gm spot. Analysis times are 5 minutes for volatile concentrations and 45 minutes for D/H ratios.