Similarity between C, N and S stable isotope profiles in European spruce forest soils: implications for the use of δ34S as a tracer

Stable isotope systematics of C, N and S were studied in soils of 5 European forest ecosystems. The sites were located along a North–South transect from Sweden to Italy (mean annual temperatures from +1.0 to +8.5 °C, atmospheric deposition from 2 to 19 kg N ha−1 a−1, and from 6 to 42 kg S ha−1 a−1). In Picea stands, the behavior of C, N and S isotopes was similar in 3 aspects: (1) assimilation favored the lighter isotopes 12C, 14N and 32S; (2) mineralization in the soil profile left in situ residues enriched in the heavier isotopes 13C, 15N and 34S; and (3) NO3–N as well as SO4–S in soil solution was isotopically lighter compared to the same species in the atmospheric input. In this study, emphasis was placed on S isotope profiles which so far have been investigated to a much lesser extent than those of C and N. Sulfate in monthly samples of atmospheric input had systematically higher δ34S ratios than total soil S at the 0–5 cm depth, on average by 4.0‰. Sulfate in the atmospheric input had higher δ34S ratios than in deep (>50 cm) lysimeter water, on average by 3.2‰. Organic S constituted more than 50% of total soil S throughout most of the profiles (0–20 cm below surface). There was a tendency to isotopically heavier organic S and lighter inorganic SO4–S, with ester SO4–S heavier than C-bonded S at 3 of the 5 sites. With an increasing depth (0 to 20 cm below surface), δ13C, δ15N and δ34S ratios of bulk soil increased on average by 0.9, 4.2 and 1.6‰, respectively, reflecting an increasing degree of mineralization of organic matter. The isotope effects of C, N and S mineralization were robust enough to exist at a variety of climate conditions and pollution levels. In the case of S, the difference between isotope composition of the upper organic-rich soil horizon (lower δ34S) and the deeper sesquioxide-rich soil horizons (higher δ34S) can be used to determine the source of SO4 in streams draining forests. This application of δ34S as a tracer of S origin was developed in the Jezeimageí catchment, Czech Republic, a highly polluted site suffering from spruce die-back. In 1996–1997, the magnitude and δ34S of atmospheric input (20 kg S ha−1 a−1, 5.8‰) and stream discharge (56 kg S ha−1 a−1, 3.5‰) was monitored. Export of S from the catchment was 3 times higher than contemporary atmospheric input. More than 50% of S in the discharge was represented by release of previously stored pollutant S from the soil. Stable isotope systematics of Jezeimageí soil S (mean of 2.5‰ in the O+A horizon, 4.8‰ in the B horizon, and 5.8‰ in the bedrock) suggests that most of the soil-derived S in discharge must come from the isotopically light organic S present in the upper soil horizon, and that mineralized organically-cycled S is mainly flushed out during the spring snowmelt. The fact that a considerable proportion of incoming S is organically cycled should be considered when predicting the time-scale of acidification reversal in spruce die-back affected areas.