Strong temporal and spatial variation of dissolved Cu isotope composition in acid mine drainage under contrasted hydrological conditions.

Copper export and mobility in acid mine drainage are difficult to understand with conventional approaches. Within this context, Cu isotopes could be a powerful tool and here we have examined the relative abundance of dissolved (<0.22 µm) Cu isotopes (δ65Cu) in the Meca River which is an outlet of the Tharsis mine, one of the largest abandoned mines of the Iberian Pyrite Belt, Spain. We followed the chemical and isotopic composition of the upstream and downstream points of the catchment during a 24-h diel cycle. Additional δ65Cu values were obtained from the tributary stream, suspended matter (>0.22 µm) and bed sediments samples. Our goals were to 1) assess Cu sources variability at the upstream point under contrasted hydrological conditions and 2) investigate the conservative vs. non conservative Cu behavior along a stream. Average δ65Cu values varied from -0.47 to -0.08‰ (n = 9) upstream and from -0.63 to -0.31‰ downstream (n = 7) demonstrating that Cu isotopes are heterogeneous over the diel cycle and along the Meca River. During dry conditions, at the upstream point of the Meca River the Cu isotopic composition was heavier which is in agreement with the preferential release of heavy isotopes during the oxidative dissolution of primary sulfides. The more negative values obtained during high water flow are explained by the contribution of soil and waste deposit weathering. Finally, a comparison of upstream vs. downstream Cu isotope composition is consistent with a conservative behavior of Cu, and isotope mass balance calculations estimate that 87% of dissolved Cu detected downstream originate from the Tharsis mine outlet. These interpretations were supported by thermodynamic modelling and sediment characterization data (X-ray diffraction, Raman Spectroscopy). Overall, based on contrasted hydrological conditions (dry vs flooded), and taking the advantage of isotope insensitivity to dilution, the present work demonstrates the efficiency of using the Cu isotopes approach for tracing sources and processes in the AMD regions.