B43A-0217:
Polychlorinated Biphenyl (PCB) Bioaccumulation in Fish: A Look at Michigan’s Upper Peninsula
B43A-0217:
Polychlorinated Biphenyl (PCB) Bioaccumulation in Fish: A Look at Michigan’s Upper Peninsula
Thursday, 18 December 2014
Abstract:
Fish consumption is an important economic, social and cultural component of Michigan’s UpperPeninsula, where safe fish consumption is threatened due to polychlorinated biphenyl (PCB)
contamination. Despite its predominantly rural nature, the Upper Peninsula has a history of industrial
PCB use. PCB congener concentrations in fish vary 50-fold in Upper Peninsula lakes. Several factors may
contribute to this high variability including local point sources, unique watershed and lake
characteristics, and food web structure. It was hypothesized that the variability in congener distributions
could be used to identify factors controlling concentrations in fish, and then to use those factors to
predict PCB contamination in fish from lakes that had not been monitored. Watershed and lake
characteristics were acquired from several databases for 16 lakes sampled in the State’s fish
contaminant survey. Species congener distributions were compared using Principal Component Analysis
(PCA) to distinguish between lakes with local vs. regional, atmospheric sources; six lakes were predicted
to have local sources and half of those have confirmed local PCB use. For lakes without local PCB
sources, PCA indicated that lake size was the primary factor influencing PCB concentrations. The EPA’s
bioaccumulation model, BASS, was used to predict PCB contamination in lakes without local sources as a
function of food web characteristics. The model was used to evaluate the hypothesis that deep,
oligotrophic lakes have longer food webs and higher PCB concentrations in top predator fish. Based on
these findings, we will develop a mechanistic watershed-lake model to predict PCB concentrations in
fish as a function of atmospheric PCB concentrations, lake size, and trophic state. Future atmospheric
concentrations, predicted by modeling potential primary and secondary emission scenarios, will be used
to predict the time horizon for safe fish consumption.