Quantifying and modeling the bioavailability of sediment-associated uranium to the freshwater midge (Chironomus dilutus)

Abstract

Uranium (U) enters aquatic environments from natural and anthropogenic sources, often accumulating in sediments to concentrations that could, if bioavailable, adversely affect benthic organisms. Current assessments of U-contaminated sediments typically rely on total U concentrations measured in the sediment, which may not be representative of the concentration of U bioavailable to benthic organisms. However, the factors and mechanisms that influence U bioavailability in sediment have not been thoroughly evaluated, despite evidence that sediment properties can influence the sorption and availability of other sediment-associated metals. The lack of detailed knowledge about the factors that modify sediment-associated U bioavailability can hinder our ability to properly use sediment quality guidelines to predict adverse effects, or lack thereof, of U-contaminated sediments to benthic communities. Therefore, the overall objective of this research was to quantify and model the key physicochemical properties of sediment that modify the bioavailability of U to a model freshwater benthic invertebrate, Chironomus dilutus. To assess the influence of sediment properties on the bioavailability of U, several 10-day sediment bioaccumulation experiments were performed exposing C. dilutus larvae to a wide range of formulated and field-collected sediments spiked with U (5, 50, 200 or 500 mg U/kg d.w.). Bioaccumulation of U in C. dilutus larvae differed by over one order of magnitude when exposed to different sediments spiked with the same total concentration of U, thus total U concentrations in the sediment displayed weak or insignificant relationships with U bioaccumulation. Concentrations of U in both the overlying water and sediment pore water collected just above and below the sediment surface displayed significant positive relationships with U bioaccumulation in C. dilutus larvae for all experiments. Significant inverse relationships were observed between the bioaccumulation of U in C. dilutus larvae and key binding properties of sediment (e.g., organic matter, fine fraction or clay, cation exchange capacity and Fe content). Simple regression equations based on the physicochemical properties of sediment successfully described the bioaccumulation of U within a factor of two and provided a significant improvement in predicting the bioaccumulation of U to C. dilutus larvae over the use of total U concentrations in the sediment. To further assess the influence of key binding properties of sediments on U bioavailability, the sorption of U to different field sediments was quantified in 48-h batch equilibrium experiments as a function of solution pH and U concentration. The degree of U adsorption to sediment was greatest at pH 6 and 7, and was significantly reduced at pH 8. The adsorption of U had a strong positive relationship with increasing binding properties of sediment up until a threshold [i.e., sediments with greater than 12% total organic carbon, 37% fine fraction (≤ 50 µm), or 29 g/kg of iron content], which generally corresponded with the observed reductions in U bioaccumulation. The data from the sorption and sediment bioaccumulation experiment were further assessed using the Windermere Humic Aqueous Model, version 7.0.4 (WHAM7), which produced reliable predictions of U sorption, total U solution concentrations, free ion concentrations and species distribution of U, which are all important for determining the risk and bioavailability of U to benthic invertebrates. More importantly, the use of WHAM7 led to successful predictions of the >20-fold difference in the U accumulation in C. dilutus larvae observed across the wide range of physicochemical characteristics of the solid and aqueous phases examined in the present studies. Overall, the research presented in this thesis quantified the significant influence of key physicochemical properties of sediment on U sorption and bioaccumulation. The use of simple regression equations and WHAM7 were able to better predict the bioaccumulation of U in C. dilutus larvae over the use of total U concentrations in the sediment. These results strongly suggest that key sediment properties such as total organic carbon, particle size distribution and iron content of sediment should be incorporated into the risk assessment and sediment quality guidelines for U in order to better predict the bioavailability of U to benthic invertebrates from U-contaminated sediments.