GSA Annual Meeting in Seattle, Washington, USA - 2017

Paper No. 69-6
Presentation Time: 9:00 AM-5:30 PM

GLASS-WATER HYDROGEN ISOTOPE FRACTIONATION IN RECENT VOLCANIC GLASSES


STAUFFER, Eric A. and CASSEL, Elizabeth J., Geological Sciences, University of Idaho, 875 Perimeter Drive MS 3022, Moscow, ID 83844, Stau2357@vandals.uidaho.edu

Volcanic glass hydrates within 103-104 years of eruption, recording the hydrogen isotope ratios (δD) of ambient meteoric water over this time interval. The δD values of hydrated volcanic glasses are now used as a proxy for δD values of ancient meteoric waters, and, accordingly, for climate and topography shortly after the time of ash deposition. Vitric volcanic rocks provide many advantages as a proxy, as ignimbrites and ash fall tuffs commonly contain readily dateable phenocrysts, can be precisely correlated, and are regionally extensive. But glass δD values are offset from that of ambient meteoric water according to glass-water hydrogen isotope fractionation, in which there is a preference for hydrogen over deuterium. Glass hydrates through both molecular diffusion and ion exchange, which differ in reaction style and proceed at different rates. Reaction rates also vary depending on temperature and glass and water compositions. But there is no glass-water fractionation factor that accounts for this variability.

Here we use samples of recent widespread ash fall tuffs, primarily the 7.7 ky Mazama ash, coupled with co-located soil water samples and in situ 12+ month soil temperature measurements to test the potential influence of temperature, chemical composition, and various location variables on hydration fractionation. Modern soil water and long-term temperature measurements are compared to post-glacial age glasses as there have not been changes in temperature, latitude, or elevation significant beyond measurement uncertainties. Soil water δD values correlate with volcanic glass δD values, and ΔδD values have an average of -22.4‰ and a range from -5.14‰ to -54.14‰. Although often applied, only one sample pair has a ΔδD value comparable to the hydration fractionation value calculated by Friedman et al. (1993). ΔδD values show no correlation with glass δD values or soil water δD values, demonstrating that ambient water composition does not control glass-water fractionation. ΔδD values do not correlate with temperature between -2.8°C to 27.6°C, demonstrating the importance of ion exchange. ΔδD values have a strong linear correlation with glass water contents, indicating that fractionation occurs consistently throughout hydration and glass composition may be an important control on glass-water fractionation.