Actinides accumulate within aquatic biota in concentrations several orders of magnitude higher than in the seawater [the concentration factor (CF)], presenting an elevated radiological and biotoxicological risk to human consumers. CFs currently vary widely for the same radionuclide and species, which limits the accuracy of the modeled radiation dose to the public through seafood consumption. We propose that CFs will show less dispersion if calculated using a time-integrated measure of the labile (bioavailable) fraction instead of a specific spot sample of bulk water. Herein, we assess recently developed configurations of the diffusive gradients in thin films (DGT) sampling technique to provide a more accurate predictor for the bioaccumulation of uranium, plutonium, and americium within the biota of the Sellafield-impacted Esk Estuary (UK). We complement DGT data with the cross-flow ultrafiltration of bulk seawater to assess the DGT-labile fraction versus the bulk concentration. Sequential elution of Fucus vesiculosis reveals preferential internalization and strong intracellular binding of less particle-reactive uranium. We find significant variations between CF values in biota calculated using a spot sample versus using DGT, which suggest an underestimation of the CF by spot sampling in some cases. We therefore recommend a revision of CF values using time-integrated bioavailability proxies.