V041-09
Tracing material transport in subduction zones: Insights from K isotopes in Izu arc lavas

Wednesday, 16 December 2020: 07:24
Virtual
Christopher A Parendo1, Stein B Jacobsen1, Jun-Ichi Kimura2 and Rex Taylor3, (1)Harvard University, Earth and Planetary Sciences, Cambridge, MA, United States, (2)JAMSTEC, Yokosuka, Japan, (3)University of Southampton, Southampton, United Kingdom
Abstract:
We report K-isotope data for recently erupted (Quaternary) lavas from the Izu volcanic arc. A primary aim of this study is to assess whether K-isotope variations can be used to disentangle the relative contributions of the different source components (i.e., mantle wedge, sediment, and altered ocean crust) that contribute material to arc lavas.

Potassium isotope analyses were made using a Nu Sapphire MC-ICPMS, utilizing He and H2 in a hexapole collision-reaction cell. External reproducibility was around 0.04‰ (2σ). Data were obtained for lavas from the frontal arc (n=24) and rear arc (n=22).

Materials from the frontal arc have values that are on the order of 0.15 to 0.35‰ heavier than the upper mantle. No significant along-arc variations are observed: volcanic islands such as Oshima, Miyakejima, Hachijojima, Aogashima, and Torishima have similarly elevated values. By contrast, most materials from the rear arc have values that are intermediate between upper-mantle and frontal-arc values. In addition, a subset of rear-arc samples show extreme variations—with values spanning a range of about 1.6‰.

We interpret the heavy K of the frontal-arc volcanoes as indicative of a dominant contribution of K by a slab-derived fluid. Strontium and Pb isotopes suggest a model in which the K is largely derived from basaltic oceanic crust, and not sediment. The heavy signature of the fluid may result from (1) extraction of K from domains of altered ocean crust with heavy K, or (2) isotopic fractionation associated with K partitioning between fluid and phases such as phengite or phlogopite. We propose that the shift to lighter K in the rear arc is in part a consequence of a larger contribution of mantle-derived K to these lavas, consistent with observed shifts in Sr and Pb isotopes. This may imply that slab-derived fluids under the rear arc exchanged material with large volumes of peridotite during transport to the locus of partial melting in the mantle wedge. Additionally, more extensive breakdown of K-bearing phases in the slab or mantle under the rear arc may cause K to partition more completely into aqueous solutions or melts, thereby suppressing isotopic fractionation effects.