SO2 disproportionation impacting hydrothermal sulfur cycling: Insights from multiple sulfur isotopes for hydrothermal fluids from the Tonga-Kermadec intraoceanic arc and the NE Lau Basin

Spatial Variations in Magmatic Volatile Influx and Fluid Boiling in the Submarine Hydrothermal Systems of Niuatahi Caldera, Tonga Rear‐Arc

“…The mean δ 34 S H2S value of the hydrothermal fluids (1.0 ± 0.6, Peters et al., 2021) and δ 34 S of their sulfide precipitates (0.7 ± 1.5 ‰, Figure 10) overlap with the assumed local host rock δ 34 S composition (1.1 ± 1.1 ‰, Keller et al., 2009), and thus suggest a S contribution by host rock leaching. The Se/S values of the bulk sulfate‐sulfide samples (1,120 ± 1,330) and pyrite (583 ± 884, Figure 5) are also consistent with host rock leaching (Figure 5).…”

“…The δ 34 S isotope composition in submarine hydrothermal sulfides is highly variable (−35 to +10 ‰, Figure 10), which is the result of S contributions from different reservoirs with distinct δ 34 S signatures, such as seawater (21.2 ± 0.9 ‰, Tostevin et al., 2014), MORB‐like mantle (−0.9 ± 0.5 ‰, Labidi et al., 2012, 2014) or subduction‐influenced mantle (5.0 ± 3.9 ‰, Alt et al., 1993; Hoog et al., 2001; Ueda & Sakai, 1984; Woodhead et al., 1987), as well as different isotope fractionation processes including magmatic SO 2 disproportion, “anhydrite buffering” and fluid‐mineral fractionation (Kusakabe et al., 2000; McDermott et al., 2015; Ohmoto & Lasaga, 1982; Ono et al., 2007; Peters et al., 2021). Although, the δ 34 S composition of the dacites hosting the hydrothermal system at Niuatahi caldera is unknown, we assume that it is comparable to the δ 34 S range of fresh volcanic rocks (1.1 ± 1.1 ‰, Keller et al., 2009) from nearby Mata volcano (Figure 1a).…”

“…Furthermore, the δ 34 S H2S value of the hydrothermal fluids and the δ 34 S of their respective sulfide precipitates strongly overlap (Figure 10) which suggests that our analyzed sulfides were near S isotopic equilibrium with the vent fluids and therefore represent source signatures and are not significantly influenced by fluid‐mineral S isotope fractionation which is typically in the range of ±1.1 ‰ (McDermott et al., 2015; Ono et al., 2007). Hence, δ 34 S values below the assumed host rock δ 34 S composition of 1.1 ± 1.1 ‰ are likely not caused by biogenic sulfate reduction or fluid‐mineral S isotope fractionation, but may be related to a magmatic volatile influx and associated magmatic SO 2 disproportionation (Figure 10, Kusakabe et al., 2000; Martin et al., 2021; McDermott et al., 2015; Peters et al., 2021).…”

“…The δ 34 S H2S of the hydrothermal fluids (1.2 ± 0.8 ‰, Peters et al, 2021) and the δ 34 S of their respective precipitates (−0.2 ± 0.5 ‰) from the Niuatahi North vent site overlap partly with the assumed host rock δ 34 S composition (1.1 ± 1.1 ‰; Keller et al, 2009) which is consistent with a host rock-derived S source (McDermott et al, 2015;Ono et al, 2007;Shanks, 2001). Low Se/S ratios in bulk sulfate-sulfide samples (38.6 ± 65.7) and in pyrite (41.2 ± 68.9, Figure 5) suggest a seawater influence (Figure 5).…”

“…Furthermore, boiling‐induced destabilization of metal (loid) complexes due to the loss of complex‐forming ligands like H 2 S is inferred to be an important process to concentrate native Au and related elements (e.g., Te, Bi) in hydrothermal fluids and their precipitates (Falkenberg et al., 2021; Gartman et al., 2018; Keith et al., 2020; Pokrovski et al., 2014). The trace element and radiogenic Pb and stable S isotope composition of hydrothermal sulfides (e.g., pyrite, chalcopyrite, and sphalerite), which precipitate during fluid boiling and/or cooling due to fluid‐seawater mixing record information about the physicochemical fluid conditions (e.g., temperature, pH, redox, salinity) and possibly about the metal (loid) sources in the crust (Fouquet & Marcoux, 1995; Keith et al., 2016; McDermott et al., 2015; Nestmeyer et al., 2021; Peters et al., 2021; Schaarschmidt et al., 2021; Wohlgemuth‐Ueberwasser et al., 2015; Yao et al., 2009). However, it is still challenging to unravel these processes based on the composition of the seafloor hydrothermal precipitates, because the source‐related signatures from the reaction zone (magmatic volatile influx and/or host rock leaching) are commonly overprinted by shallower hydrothermal processes (fluid boiling and/or seawater mixing).…”