To date, most estimates of contaminant fluxes across the sediment/water interface in risk assessments have been done using diffusive flux models. However, the reliability of these is limited as the overall flux from the sediment may have contributions caused by advection and bioturbation. We found through a comparison of modelled fluxes versus measured fluxes, that the methods Benthic Flux Chamber and surface leaching tests in a risk assessment context showed similar magnitude while calculated fluxes deviated at least by a factor of 100 from measured fluxes. This may be explained by the flux contribution in connection with bioturbation. The chamber- measured fluxes of copper were low compared to those of zinc and cobalt, but this is consistent with leaching tests that indicated copper to be more strongly bound. Risk assessments based on total concentrations may be misleading.