Significance of the terrestrial sink in the biogeochemical sulfur cycle

Sample collection and preparation Samples of stream and lake water, snow and glacier meltwater, fluvial sediment, bedrock and soil were collected in July 2015 along the forefield of Jostedal Glacier, Norway, and water temperature and pH were measured during sampling. Sediment samples were wet-sieved to separate gravel (>2 mm), sand (2 mm - 62.5 µm), and mud (<62.5 µm) fractions. The fine-grained (<62.5 µm) sediments were reacted with buffered acetic acid (pH = 4.8) and hydrogen peroxide to remove carbonate and organic matter, respectively, and freeze-dried. Considering the wide variety of lithology within the igneous and metamorphic bodies in the area, the sands from the most proximal part of the stream were further sieved to isolate the medium sand (250-500 µm) fraction, which we used as a representative of the “average bedrock” in the drainage of the Storelvi River for the measurement of major element concentrations. Basement rock samples were used for concentration and isotope ratio of sulfur, since the exposure to weathering and depositional processes can result in loss of sulfur via oxidation. The bedrock samples, soil samples, and fine-grained (<62.5 µm) sediment samples were freeze dried and powdered for further analyses. Liquid sample chemistry Cation and anion concentrations in stream water were quantified by Dionex DX 500 ion chromatography equipped with both cation-exchange column CS-12 and anion-exchange column AS-11. Sulfur isotope analysis Sulfate in stream water, snow, and glacier meltwater was purified and concentrated by anion chromatography on Bio-Rad AGX8 anion exchange resin and analyzed for 34S/32S ratios using a Neptune Plus multi-collector inductively coupled plasma mass spectrometer at Caltech. The measured 34S/32S ratio was calibrated using a linear interpolation between the two bracketing standard values, and sulfur isotope compositions were reported using conventional delta notation with respect to VCDT (Vienna Cañon Diablo Troilite). The bracketing standard was calibrated against the IAEA S-1 reference material and has a δ34S value of -1.55‰ ± 0.16. Analytic reproducibility for δ34S has been evaluated as ± 0.2‰. Total sulfur in bedrock samples was reduced to H2S by a boiling mixture of HI-HCl-H3PO2 (Thode et al., 1961) and precipitated as Ag2S. Freeze-dried and powdered soil samples were fused with Eschka’s mixture (MgO and CaCO3) at 800 °C, converting available sulfur to BaSO4 (Selvig & Fieldner, 1927). Sulfur isotope ratios of recovered Ag2S and BaSO4 were determined using a Delta V Plus continuous-flow isotope ratio mass spectrometer operated by the Department of Earth and Planetary Sciences at Northwestern University. All measured sulfur isotope compositions were reported in the delta (δ) notation as isotope ratio variations relative to VCDT, and linear scale correction was applied to yield δ34S values of 0.5‰ and 20.3‰ for IAEA SO-5 and NBS-127 (BaSO4), and -0.3‰, 22.7‰, and -32.3‰ for IAEA S-1, S-2, and S-3 (Ag2S). Analytical precision is within 0.2‰. Solid chemistry The “average bedrock” and fine-grained sediment samples were analyzed for major elements by inductively coupled plasma atomic emission spectroscopy.