Cordilleran Section - 119th Annual Meeting - 2023

Paper No. 1-11
Presentation Time: 11:40 AM

HYDROGEN ISOTOPE SIGNATURES OF MAGMA ASCENT AND CO2 FLUXING IN OLIVINE-HOSTED MELT INCLUSIONS


PAMUKCU, Ayla, Earth and Planetary Sciences, Stanford University, 450 Jane Stanford Way, Stanford, CA 94305-2004 and GAETANI, Glenn, Geology and Geophysics, Woods Hole Oceanographic Institution, Woods Hole, MA 02453

Ascent-driven degassing and CO2 fluxing are two processes often invoked to explain melt inclusion (MI) suites whose H2O and CO2 contents do not conform to degassing paths. Both of these processes dehydrate the host melt, providing the driving force for diffusive water loss from MIs. Water loss and related formation of a CO2-rich vapor bubble (VB) causes MI compositions to move to both lower H2O and CO2 values. Consequently, both processes can produce similar scatter on H2O-CO2 plots, so that discerning which process was responsible is not possible on the basis of volatile contents alone. Experiments on olivine-hosted MIs show that H2O can diffuse rapidly through olivine, and that deuterium (D) diffuses more slowly than hydrogen (H). We compare hydrogen isotope fractionation induced by degassing and CO2 fluxing in olivine-hosted MIs to assess if hydrogen isotopes can be used to distinguish the cause of water loss. We examine a MI suite from Hut Point Peninsula (Antarctica), whose chemistry (H2O content and D/H values) is consistent with diffusive water loss. We use a spherical geometry with radial diffusion to model equilibration that includes generation of a VB during waterloss. The path of dissolved H2O and D/H in the external melt differs for ascent and fluxing. For ascent models, we calculate the external melt H2O path using MagmaSat and allow the external melt D/H to become lighter with degassing. For fluxing models, we estimate the external melt H2O path using isobaric, isenthalpic CO­2 fluxing simulations run in ENKI and hold the external melt D/H value constant. We find that fluxing-induced fractionation is mild (~6‰ max) and limited to the initial part of the simulation when hydrogen outpaces deuterium diffusion. The model ultimately returns to the external melt value. Ascent produces considerable fractionation over the same timescale (~20-50 ‰), but the onset of fractionation is a strong function of ascent rate. Our results suggest that the D/H fractionation and water loss signature of the Hut Point MIs are more likely a record of ascent than fluxing.