Les métaux échantillonnés par la DGT sont-ils représentatifs de la fraction biodisponible ?

[Notes_IRSTEA]Présentation PowerPoint de 14 diapositives [Departement_IRSTEA]Eaux [TR1_IRSTEA]BELCAInternational audienceBioavailability of trace metals for aquatic organisms in the dissolved compartment of water is usually interpreted with the Free Ion Activity Model (FIAM) and its extension, the Biotic Ligand Model (BLM). Under fame generally hypothesis, the FIAM states that the biological response to a metal exposure is proportional to the activity of the free-ion {Mz+} in solution. The BLM generalizes this concept and includes competition between metal and major cations (Ca2+, Mg2+, Na+, K+ and H+) on specific binding sites at the surface of the biological membrane. As a speciation device, DGT samples labile metal concentration: free ion plus a fraction of organic metal complexes. It is claimed to estimate bioavailable fractions of metals. The aim of this communication is to synthesize the results of some recent work comparing biological response and labile fraction sampled by DGT in order to evaluate the ability of DGT to assess bioavailability. We investigated two types of biological responses: acute toxicity and bioaccumulation. In the first part, DGT were used to measure metals inducing lethality on daphnia magna during acute toxicity tests. Various media were studied: synthetic organic ligands, humic acids, and natural organic matter (algae and river extracts, wastewater), with two types of gels (classical open pores and restrictive pores). Copper and cadmium were studied. Measurements by DGT equipped with restricted gels appear to be in good agreement with inorganic fraction of metal, even if labile fraction could overestimate bioavailable fraction. In the second part, the accumulation of copper by aquatic mosses at environmentally relevant concentrations (1-5 µg.L-1) and with various organic ligands is compared to DGT-labile restricted gels measurements. In this case, DGT tends to underestimate the bioavailable fraction. This suggests that not only inorganic copper but also some weak organic complexes are available for aquatic mosses at realistic metal concentrations. Finally, we discuss the role of major cations in the bioavailability of copper, in the cases of acute toxicity and bioaccumulation at realistic environmental concentrations. Ca2+ and Mg2+ compete indeed with copper and play a protective role, which would not be highlighted by the sampling with DGT