Sulfur isotopic variations in the products of the 1895 CE eruption at Zao volcano (NE Japan): Implications for connecting eruption source and syn-eruptive magmatic-hydrothermal processes

Abstract

Sulfur isotopic ratio in sulfate and sulfide in subvolcanic hydrothermal systems is a valuable tracer to study the magmatic-hydrothermal processes from the magma source through to volcanic eruptions. Zao volcano is among the most active volcanoes in NE Japan, with historical explosive eruptions occurring during the last thousand years and unrest episodes since 2013. This necessitates a detailed assessment of the potential risk of future volcanic hazards. We investigated the magmatic-hydrothermal processes that occurred during the 1895 CE eruption sequence at Zao volcano by conducting mineralogical and sulfur isotope analyses in the exposed well: (i) six volcanic units (Layers 1–6) of the 1895 CE eruption products (clayish ash deposits with andesitic bombs, lapilli of scoria, and minor altered lithic fragments) deposited on the rim of Okama crater lake; and (ii) clay-altered and silicified rocks from the Nigorikawa alteration zone (NGA) surrounding the Goshikidake cone. Mineralogical data show that the samples mainly consist of alunite, pyrite, and gypsum. Alunite and pyrite occur as fine crystal mixtures associated with mineral assemblages of both advanced argillic alteration (i.e., those of cristobalite and kaolinite) and silicification (i.e., those of cristobalite, tridymite and native sulfur). Gypsum typically appears as isolated euhedral crystals of several millimeters in size. Samples of the 1895 CE eruption products have a narrow range of δ34S values from +3 ‰ to +5 ‰ for gypsum, from +9 ‰ to +13 ‰ for alunite, and approximately −10 ‰ for pyrite. For the NGA samples, the δ34Sgypsum, δ34Snative sulfur, and δ34Spyrite values range from −12 ‰ to −9 ‰, whereas for alunite, these range from +8 ‰ to +18 ‰. This indicates that alunite and pyrite in the 1895 CE eruption products were derived from the advanced argillic alteration and silicification zones that developed under Okama crater, which is exposed as the NGA. Estimated alteration temperatures based on the sulfur isotopic equilibrium between alunite and pyrite pairs are 200 °C–300 °C. By contrast, δ34Sgypsum values in the 1895 CE products are significantly higher than those in the NGA (which are derived from oxidation of pyrite or H2S, or both), ranging between an estimated parental fluid of δ34Sbulk-initial = ca. +1 ‰ and the Quaternary volcanic rocks of the Japan arc. This suggests that gypsum in the 1895 CE eruption products derived from magmatic vapor condensate (anhydrite) formed in the volcanic conduit during the eruption, thus becoming replacement of anhydrite by gypsum after or during the tephra deposition on the Zao summit surface. Our results on sulfur-bearing minerals provide new clues for better understanding (and monitoring) the syn-eruptive processes of volcanic eruptions focused on subvolcanic hydrothermal systems.