Characterisation of the deposition and erosion fluxes of sediments: Experimental protocol and multi-class modelling.

El Ganaoui O., Schaaff E.*, Boyer P., Amielh M.**, Anselmet F.** and Grenz C.* Institut de Radioprotection et de Sûreté Nucléaire ; CE - Cadarache Bat 159 ; 13108 St-Paul Lez Durance, France *Laboratoire d'Océanographie et de Biogéochimie, Centre d'Océanologie de Marseille Station Marine d'Endoume, Rue de la Batterie des Lions, 13007 Marseille **Institut de Recherche sur les Phénomènes Hors Equilibre, Technopôle Château-Gombert, 49 rue F. Joliot-Curie, B.P.146 13384 Marseille Cedex 13, France Bioflow First Workshop / 10-14 September at NIOO-CEMO, Yerseke, the Netherlands

Toxic waste released in the hydrosystems by industrial facilities interacts with water, suspended sediment and bottom sediments by different processes e.g. dispersion, adsorption and mass exchange between suspended and bottom sediments. This later exchange is conditioned by sediment dynamics characterised by erosion and deposition fluxes. Consequently, assessing pollutant transfers in rivers, lakes and estuaries, requires an evaluation of the transport and fate of the sediments. Pollutants (metals, radionuclides…) mainly react with cohesive sediments. The application of models used to evaluate the behaviour of these elements requires the description of ‘erosional’ properties of the bed, such as the critical shear stress of erosion, the erosion rate, the critical shear stress of deposition, and the settling velocity. In practice, no theory is yet available for the calculation of these properties so direct in-situ measurements or through the use of an experimental flume are needed. Moreover, a natural water flow contains a wide and continuous range of granulometric classes of particles of different types, each one being characterised by specific dynamics and, depending on the physical conditions, a specific ability to absorb pollutants. In this context, we will present an experimental protocol to identify the main sedimentary classes and assess their specific dynamic parameters. The recirculating flume equipment and the experimental protocol will be described, results acquired from river, estuarine and continental shelf sediments will be presented and the analytical method, based on a multi-class model suited to the flume will be discussed. Comparison between this approach and commonly used models without any particle class distinction, points out the efficiency of the multi-class modelling of estuarine and/or freshwater bottom sediment behaviour.