Spatial and temporal variability of PCBs in Detroit River water assessed using a long term biomonitoring program.

An assessment of polychlorinated biphenyls (PCBs) in water of the Detroit River, North America, was performed using mussel biomonitoring data generated between 1996 and 2010. The study included a temporal monitoring program at six locations and an enhanced spatial survey performed during 2002. Mussels were transplanted at biomonitoring stations and collected after 21-226 d. A toxicokinetic model was used to perform steady state and control corrections followed by an equilibrium partitioning model to estimate bioavailable water concentrations of PCBs. Bioavailable water concentrations of PCBs were similar in magnitude and trends to dissolved water concentrations reported in previous studies using direct sampling approaches. PCB concentrations exhibited complex temporal patterns at the six biomonitoring stations with multi-year declines in PCB trajectories between 1996 and 2002 and less consistent trends occurring across stations in later years. Spatial patterns of PCBs during 2002 revealed significantly higher water contamination on the US side of the river (seasonal average mean ± standard error (SE) sum PCB concentration of 0.63 ± 0.11 ng L(-1)) compared to the Canadian side (mean ± SE sum PCB concentration of 0.09 ± 0.01 ng L(-1)). Spatial/temporal variability of bioavailable PCB concentrations was greatest between countries, followed by moderate variation across years and by river reach (upstream, midstream and downstream sections within a country) and lowest when comparing intra-seasonal variation.

Determination of in situ polychlorinated biphenyl elimination rate coefficients in the freshwater mussel biomonitor Elliptio complanata deployed in the Huron-Erie corridor, southeast Michigan, USA, and southwest Ontario, Canada.

Quantitative mussel biomonitoring studies often use laboratory-measured whole-body elimination rate coefficients (k(tot) in conjunction with concentration data from field-deployed mussels to estimate bioavailable steady-state contaminant concentrations (C(m(ss)) in aquatic systems. However, the validity of applying laboratory k(tot) values to field-deployed mussels has not been verified. The present study quantified in situ k(tot) values by the common mussel biomonitor species Elliptio complanata for a group of performance reference compound polychlorinated biphenyl (PCB) congeners (International Union of Pure and Applied Chemistry 23, 61, 109, and 173) at 11 sites along the Lake Huron-Lake Erie corridor (southeast MI, USA, and southwest ON, Canada). Predictive site-specific k(tot) versus log K(ow) relationships were derived to estimate in situ k(tot) values for the bioaccumulated environmental PCB congeners, and the resultant steady-state concentration estimates were compared against values estimated using laboratory-derived k(tot) values. In situ k(tot) values were almost always faster (1.3-10-fold) than rate coefficients determined by laboratory studies, with a mean difference +/- standard error of 3.89 +/- 0.28-fold. Consequently, control adjusted steady-state concentrations of sum PCBs in the mussels were overestimated by factors ranging from 1.30 to 3.45 across sites when using laboratory-derived k(tot) values in place of in situ k(tot) values and a field deployment period of 90 d. This error was shown to increase with increasing congener hydrophobicity, as represented by their log K(ow) values, and when shorter field deployment times are performed. To more accurately predict the steady-state concentrations of environmentally accumulated PCBs in mussels, it is recommended that future quantitative biomonitoring studies employ the reference compound approach for measuring site-specific in situ k(tot) values.