Metal accumulation risks in regularly flooded and non-flooded parts of floodplains of the River Rhine: Extractability and exposure through the food chain

Abstract

Ecotoxicological risks of sediment contamination in floodplains are supposed to be highest in the regularly flooded parts. Therefore, in risk assessments, the non-flooded parts are neglected or considered to be reference areas. We investigated the metal extractability and levels in important food sources for vertebrates, viz. grass shoots and earthworms, in flooded as well as non-flooded parts and compared these with total metal concentrations. A comparison of these areas in the moderately polluted ‘Afferdensche en Deestsche Waarden’ floodplains along the River Rhine showed that total Zn, Pb, and Cd concentrations were highest in the regularly flooded parts. However, CaCl₂-extractable Zn concentrations were highest in non-flooded areas, and those of Pb and Cd were equal in both areas. Total Cu concentrations were not significantly different between the two areas, but CaCl₂-extractable Cu concentrations were highest in the regularly flooded areas. The metal concentrations in grass shoots of non-flooded areas were equal to (Zn, Cu, Cd) or higher than (Pb) those in regularly flooded areas. Zn concentrations in earthworms in regularly flooded areas were higher, but concentrations of Cu, Pb, and Cd were not. Ecotoxicological risk assessments require analysis of the total and potentially bioavailable metal concentrations in soils as well as concentrations in biota. This study shows that the less contaminated non-flooded areas in moderately polluted floodplains cannot be neglected in metal accumulation studies and cannot be used as pristine reference areas.

Keywords:

• Contaminated floodplains
• Metal accumulation
• Non-flooded areas
• Food web
• Bioavailability; Risk assessment

Acknowledgements

We would like to thank Franciska de Vries, Erika Koelemij, Gerrit Jungheim, and Mathilde Zorn, for their valuable help with the sampling and analyses, Jelle Eygensteyn, Liesbeth Pierson, and Rien van der Gaag, for their assistance with the ICP analyses, Martijn Dorenbosch, for assistance during statistical analyses, Gertjan Geerling, for assistance with ArcMap, Jan Klerkx, for his linguistic comments, and the State Forestry Services (SBB), for their permission to do research in the ADW floodplain. This research is part of the NWO-SSEO and the NWO-LOICZ programme and is CWE publication no. 393.