The long-term management of contaminated dredged sediments raises the problem of the fate of the potentially toxic element contained in these matrixes. The physico-chemical parameters influence the speciation and distribution of contaminants between organic or inorganic bearing phases, and the terrestrial management of sediment induces the modification of factors such as oxydation, wetting / drying cycles or bacterial activity that will influence the physico-chemical parameters and thus the contaminant speciation. In order to identify the main mechanisms responsible for the mobility ofpotentially toxic elements and to estimate the environmental acceptability of dredged sediment in valuation scenario (as hill landscape, road fill or undercoat), the study was organized following three main steps I characterizing the sediment (particle size, mineralogy, moisture content, solid phase and pore water composition) and evaluating factors (pH, L / S, temperature ...) that control the leaching of elements, according to standardized procedures, II developing a set of input parameters for the geochemical code ORCHESTRA according to standardized procedures (quantification of the most reactive carrier phases : clays, carbonates, oxyhydroxides of iron or aluminum and organic matter - fulvic and humic acid) III modeling and predicting the solubility curves of the elements described in the standardized tests from step (i) using the set of input parameters defined in step (ii). Leaching tests and implementation models are complementary approaches that are necessary to understand the mechanisms controlling the mobility and retention of elements. Modeling of column dynamic leaching tests is very sensitive to changes in input parameters, so the model for the major elements should be as adequate as possible. The obtained predictions for Al, Ca, Cl, Fe, H2CO3, Mg, Si, SO4 2-, Cu, Cr, MoO4 2-, Pb and Zn were close to the experimental data, which indicates that the main processes controlling the solubility of elements were taken into account. The predictions for Ni and As, however, were not satisfactory, showing that some retention processes are still unknown or were not taken into account by the database MINETEQ2A. A better description of As behavior would require to include inAs complexation by the MON in the module Nica-Donan.
