V11C-4736:
Variations in Magmatic He, CO2 and δ13 C during the 2005-06 Seafloor Eruption Near 9°50'N on the East Pacific Rise

Monday, 15 December 2014
David W Graham, Oregon State University, College of Earth, Ocean, & Atmospheric Sciences, Corvallis, OR, United States, Peter J Michael, University of Tulsa, Tulsa, OK, United States and Kenneth Howard Rubin, Univ Hawaii, Honolulu, HI, United States
Abstract:
We report dissolved (glass) CO2 concentrations, plus vesicle He and C isotope compositions and He and CO2 abundances, for a suite of 23 basalt glasses from the 2005-2006 eruption on the East Pacific Rise. This sample suite provides a unique opportunity to study magma recharge along the mid-ocean ridge, to quantify degassing prior to and during eruption, and to evaluate the extent to which mantle carbon is incorporated into hydrothermal fluids and ultimately into seafloor ecosystems. Our work covers the full spatial and temporal extent of the months-long eruption (ages from 210Po-210Pb dating), and is complementary to a previous study of volatile changes with distance from the vent during part of the eruptive episode (Soule et al. 2013; EPSL). Vesicle 3He/4He is uniform at 8.56±0.04(1σ) RA. Vesicle He and CO2 abundances vary by factors of 17 and 230, respectively. The variability in CO2/He ratio is evidence for kinetic fractionation of He from CO2 during vesiculation, and reveals the potential for using vesicle CO2/He as a relative geo-speedometer for the rate of magma ascent and lava emplacement. Nearly all samples lie in a narrow range of 280-370 ppm for total CO2 (vesicles+glass) suggesting saturation at a depth of ~1.5 km in the crust, similar to the depth of the seismically imaged magma lens. Vesicle CO2/He co-varies positively with the fraction of CO2 contained in vesicles (r2=0.94) and negatively with vesicle δ13C (r2=0.83). The δ13C ranges from -2.6 to -5 per mil, and straddles the value for CO2 in EPR vent fluids (M. Lilley, pers. comm.). The C isotope results are consistent with vapor phase 13C-enrichment during CO2 exsolution from a basaltic melt, as determined experimentally (Javoy et al. 1978; Contrib Mineral Petrol). Collectively the observations are explained by quasi-closed-system degassing, where melt ascent from the magma lens was sufficiently rapid that minimal bubble loss occurred during the eruption and flow of lava onto the seafloor.