Importance of growth rate on mercury and polychlorinated biphenyl bioaccumulation in fish

To evaluate the effect of fish growth on mercury (Hg) and polychlorinated biphenyl (PCB) bioaccumulation, a non-steady-state toxicokinetic model, combined with a Wisconsin bioenergetics model, was developed to simulate Hg and PCB bioaccumulation in bluegill (Lepomis macrochirus). The model was validated by comparing observed with predicted Hg and PCB 180 concentrations across 5 age classes from 5 different waterbodies across North America. The non-steady-state model generated accurate predictions for Hg and PCB bioaccumulation in 3 of 5 waterbodies: Apsey Lake (ON, Canada), Sharbot Lake (ON, Canada), and Stonelick Lake (OH, USA). The poor performance of the model for the Detroit River (MI, USA/ON, Canada) and Lake Hartwell (GA/SC, USA), which are 2 well-known contaminated sites with possibly high heterogeneity in spatial contamination, was attributed to changes in feeding behavior and/or prey contamination. Model simulations indicate that growth dilution is a major component of contaminant bioaccumulation patterns in fish, especially during early life stages, and was predicted to be more important for hydrophobic PCBs than for Hg. Simulations that considered tissue-specific growth provided some improvement in model performance particularly for PCBs in fish populations that exhibited changes in their whole-body lipid content with age. Higher variation in lipid growth compared with that of lean dry protein was also observed between different bluegill populations, which partially explains the greater variation in PCB bioaccumulation slopes compared with Hg across sampling sites. Environ Toxicol Chem 2018;37:1655-1667. © 2018 SETAC.