Assessing sediment hazard through a weight of evidence approach with bioindicator organisms: a practical model to elaborate data from sediment chemistry, bioavailability, biomarkers and ecotoxicological bioassays

Quality assessments are crucial to all activities related to removal and management of sediments. Following a multidisciplinary, weight of evidence approach, a new model is presented here for comprehensive assessment of hazards associated to polluted sediments. The lines of evidence considered were sediment chemistry, assessment of bioavailability, sub-lethal effects on biomarkers, and ecotoxicological bioassays. A conceptual and software-assisted model was developed with logical flow-charts elaborating results from each line of evidence on the basis of several chemical and biological parameters, normative guidelines or scientific evidence; the data are thus summarized into four specific synthetic indices, before their integration into an overall sediment hazard evaluation. This model was validated using European eels (Anguilla anguilla) as the bioindicator species, exposed under laboratory conditions to sediments from an industrial site, and caged under field conditions in two harbour areas. The concentrations of aliphatic hydrocarbons, polycyclic aromatic hydrocarbons and trace metals were much higher in the industrial compared to harbour sediments, and accordingly the bioaccumulation in liver and gills of exposed eels showed marked differences between conditions seen. Among biomarkers, significant variations were observed for cytochrome P450-related responses, oxidative stress biomarkers, lysosomal stability and genotoxic effects; the overall elaboration of these data, as those of standard ecotoxicological bioassays with bacteria, algae and copepods, confirmed a higher level of biological hazard for industrial sediments. Based on comparisons with expert judgment, the model presented efficiently discriminates between the various conditions, both as individual modules and as an integrated final evaluation, and it appears to be a powerful tool to support more complex processes of environmental risk assessment.