The present study aims to evaluate and verify toxicokinetic models for the bioaccumulation of Cd, Pb, Cu and Zn in the gammaridean amphipod Gammarus zaddachi (Sexton 1912) from the River Hunte (Germany). The bioaccumulation experiment was performed in a static system, taking into account the effect of body size on bioaccumulation and the relationship between toxicokinetic model parameters and the permeable body surface area of gammarids. A modified two-compartment model was employed, which was not limited to changes in both biomass of gammarids in the experiments and metal exposure concentrations; the result was a significant model fit. The parameters k(1) and BCF decreased with increasing body length (BL) of G. zaddachi, while no such trend was observed for k(2) among BLs ranging from 8.1 to 24.1 mm. The nonlinear relationship was successfully quantified using an inverse-sigmoid logistic model as a basis for subsequent adjustment of the toxicokinetic uptake models. Moreover, k(1) and BCF increased with increasing specific surface area (SSA, i.e., the ratio of permeable body surface area to body volume) of gammarids. This relationship was successfully quantified using an extended Langmuir equation, which was derived by combining the inversely sigmoid logistic relationship between k(1), BCF(Exp) and BL with a single-exponential-decay relationship between SSA and BL reported previously, implying that the size-dependent bioconcentration was dominated by the SSA-related uptake. Thus, with increasing SSA, k(1) and BCF at first increased sharply, for smaller SSAs with larger BLs, but then approached saturation for larger SSAs with smaller BLs. In addition, field-to-experimental BCF ratios (BCF(Field)/BCF(Exp)) were determined, yielding values around 1 (indicating equality of the two BCFs) for Pb and Cd for smaller amphipods, but much higher values for larger sizes. The BCF ratios for Cu and Zn were much larger than 1 for both smaller and larger sizes. However, when seasonal changes in BL distribution of gammarids were considered, no significant differences were observed between annual ranges of BCF(Field) and BCF(Exp) for Pb and Cd. Considering the seasonal changes in BL distribution as well as the Cu and Zn metabolic requirements, no significant difference was observed for Cu, but still a significant one for Zn. From an ecotoxicological perspective we suggest that in the verification of toxicokinetic models not only field-to-experimental BCF ratios should be taken into account, but also several ecological factors such as the size distribution of the animal populations under study as well as, if applicable, metabolic requirements for essential elements.
