d³⁴S_SO4, d¹⁸O_SO4, ³⁵S_SO4 and solute chemistry were measured in rain, snow, spring water, and stream water in Loch Vale, an alpine basin in the Colorado Rocky Mountains, during 1999 to investigate sources, cycling, and residence times of sulfur (S).  The primary source of S in Loch Vale and many other high-elevation, granitic basins in the Rocky Mountains is atmospheric deposition of SO₄.  Sulfur mass balance calculations for Loch Vale indicate that S exports exceed S inputs by 20 to 50%, suggesting that sources of S other than atmospheric deposition contribute to stream fluxes of SO₄.  Pyrite occurs in the bedrock in trace quantities, however, only one specimen has been found in Loch Vale during two decades of study.  Another possible source of the “excess” S is a net export of S stored in soils and wetland peat that may have accumulated when SO₄ deposition was higher (1960s – 80s) or under different climate conditions.

d³⁴S_SO4 was used in a binary mixing equation to quantify the proportion of S in spring and stream water derived from atmospheric deposition and from terrestrial sources.  The average d³⁴S_SO4 of precipitation in Loch Vale is 5.6±1.0‰ (n=15).  d³⁴S_SO4 was regressed against 1/SO₄ and the y-intercept was taken to represent the d³⁴S_SO4 of pyrite.  Results indicated that 62±16% of annual SO₄ flux in the spring and 42±7% of annual SO₄ flux in the stream were derived from atmospheric deposition.  Regression equations for the spring and stream had y-intercepts of 1.5 and –1.5, respectively.  The difference in y-intercepts could be due to differences in d³⁴S_SO4 of pyrite in the stream and spring basins, or to differences in the importance of biologically-mediated S cycling.

The d¹⁸O_SO4 of spring and stream waters (-5.0 to 2.1) were substantially lighter than that of atmospheric deposition (13±1‰), consistent with the hypothesis that atmospherically deposited SO₄ is biologically cycled and/or mixed with S from mineral weathering.  Radiogenic ³⁵S_SO4, which is deposited from the atmosphere and has a half-life of 87 days, was used to calculate the fraction of “new” SO₄ in spring and stream water.  Results indicated that “new” SO₄ accounted for 28±7% and 32±8% of annual flux in the spring and stream, respectively.  Comparison with d³⁴S_SO4 results indicates much of the atmospherically-deposited SO₄ is stored in the watershed for more than one year.