The effects of soil leaching on metal bioavailability, toxicity and bioaccumulation in «hordeum vulgare» cultivated in copper amended soils

Discrepancies in ecotoxicity effects data derived from soils amended with dissolved metals in the laboratory and soils from historically aged metal-contaminated sites pose a challenge in deriving environmentally relevant soil quality criteria. In this thesis, the chemical artefacts produced by dissolved metal additions and the subsequent effects on metal bioavailability, toxicity and bioaccumulation are examined. The overall goal of this research is to advance the understanding of soil trace metal bioavailability mechanisms and improve current methods for ecotoxicity testing and bioavailability modeling. The first objective was to develop a leaching method which would minimize the "salt-effect" commonly observed in freshly-spiked soil samples. This was achieved by adapting a column leaching procedure in order to spike and leach larger volumes of test soil as a preparatory step prior to ecotoxicity assays. An experiment was set up to observe changes to leachate chemistry resulting from the dissolved Cu additions and subsequent leaching with two weak electrolyte solutions. Results showed that leaching removed the excess dissolved Cu as well as the excess acidity and base cations that were solubilized during the Cu additions. Differences between control and spiked samples of some key soil solution parameters (e.g. pH, DOC and total dissolved Ca, Mg, Al, Fe) were reduced thereby resulting in a set of spiked samples more conducive to deriving causal dose-response relationships. A second experiment compared soil and soil solution properties of samples which underwent the spike/leach procedure to those of freshly spiked samples. Lower soil pH (up to 0.81 pH units) and DOC concentrations were observed in non-leached samples as well as up to 35-fold and 55-fold increases of dissolved Ca and Mg concentrations, typical of the salt-effect. It was estimated that the non-leached samples contained up to 100-fold greater Cu2+ and 50-fold greater Al3+ activities which, I hypothesized, could result in greater phytotoxicity of non-leached samples. Bioassays were conducted wherein Hordeum vulgare (barley) seedlings were exposed to the leached and non-leached Cu-spiked soil samples for 14 days. The leached samples were less toxic to barley and showed significant increases (up to 1.7-fold) in median inhibitory concentrations (i.e. IC50) for root elongation in two of the three test soils. The Cu2+ fraction was able to explain much of the variability in toxicity between leached and non-leached samples, as well as among the different test soils. One exception was the most acidic test soil for which plants in the leached samples showed up to 10-fold greater toxicity than plants in the non-leached samples, when exposed to similar Cu2+ activities. Soil speciation and bioaccumulation data for Cu, Al and Ca were used to deduce that Ca deficiency and possibly Al toxicity contributed to the toxicity observed in this sample set. In a modeling exercise, the concepts of the terrestrial biotic ligand model (TBLM), which take into account ion competition at the soil solution/root interface, to describe toxic response in our samples. Results showed that the additive effects of Cu and Al toxicity estimated from the Cu-root ligand complex (Cu-BL+) and Al-root ligand complex (Al-BL2+) accounted for more of the variability in toxic response data than did the Cu2+ fraction or the Cu-BL+ fraction alone. The model however, could not account for the effects of the Cu-induced Ca deficiency in the most acidic test soil, thus highlighting one of the limitations of the TBLM. This thesis has contributed to the understanding of metal bioavailability mechanisms affected by soil spiking procedures and highlights the benefits of soil leaching in preparing soil samples for ecotoxicity assays which should improve trace metal risk assessment in the future.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.