Abstract
This study reports radiogenic Sr isotope ratios (Sr-87/Sr-86), stable Sr isotope ratios (delta Sr-88/86), and major ion concentrations for river, rock, sediment, soil, and plant samples collected from the Cleddau and Hollyford catchments in the Milford Sound region of Fiordland, New Zealand. The catchments primarily drain gabbro, but some tributaries access limestone and volcanogenic sediments. The goal of the study was to understand controls on riverine delta Sr-88/86 values in a landscape with multiple factors that may influence chemical weathering, including dense vegetation, high rainfall, and abundant, freshlyeroded Holocene fluvio-glacial and landslide debris.
Rivers draining gabbro have higher delta Sr-88/86 values than bedrock, by as much as similar to 0.14%, and the delta Sr-88/86 values strongly correlate with molar Ca/Sr ratios (R-2 = 0.69). Leaching of rocks and sediment reveals no evidence for the preferential dissolution of minerals having high delta Sr-88/86 values and Ca/Sr ratios. In-stream Sr isotope fractionation seems unlikely because comparison against Sr-87/Sr-86 and Ca/Sr ratios demonstrates that riverine delta Sr-88/86 values conservatively trace water-mass mixing. The riverine data are best explained by the input of soil water, which is distinct from potential bedrock end-members (i.e., silicates and carbonates) based on delta Sr-88/86 but indistinguishable in terms of Ca/Sr and Sr-87/Sr-86. While strontium isotope fractionation during secondary mineral formation and pedogenesis is possible, clay mineral formation is minor and most soils are poorly developed. Instead, soil water delta Sr-88/86 values more likely reflect plant uptake. Plant samples yielded a wide range of delta Sr-88/86 values, but on average, they are lower than those for bedrock, consistent with the expectation that plants preferentially incorporate lighter Sr isotopes. Mass-balance constraints, together with Sr-87/Sr-86 ratios, indicate that soil water delta Sr-88/86 values are similar to 0.30% higher than bedrock delta Sr-88/86 values, and mixing calculations show that the plant-fractionated soil water pool contributes similar to 27% of the riverine Sr. For tributaries accessing limestone and volcanogenic sediments, Ca/Sr and Sr-87/Sr-86 ratios appear consistent with two-component mixing between silicate and carbonate weathering, but delta Sr-88/86 values reveal a third contribution from soil water inputs, similar to gabbro catchments.
The results of this study suggest that Sr isotopes behave conservatively during water mass mixing and stream transport but non-conservatively in soil, where plant uptake can elevate soil water delta Sr-88/86 values relative to bedrock. Plant uptake, or related biogeochemical processes, such as ion-exchange on organic matter surfaces, also appear to modify soil water Ca/Sr ratios. Many studies use Sr-87/Sr-86 and Ca/Sr ratios to apportion riverine solutes between silicate and carbonate weathering, but Ca/Sr ratios may be non-conservative in densely vegetated areas. The stable Sr isotope tracer shows promise for resolving riverine cation sources, as well as effects from biological cycling. (C) 2015 Elsevier Ltd. All rights reserved.