| Increasing awareness about the presence and ecological ramifications of toxic, persistent and bioaccumulative contaminants within the Great Lakes system spurred the implementation of numerous bans and emissions restrictions over the past few decades. Due to their high trophic status in food webs and the critical link they serve with human consumers, fish species have historically been monitored to assess the relative success of such remedial efforts within the region and to simultaneously ascertain the current risks posed to local humans. Using Bayesian dynamic linear modelling, this project first aimed to evaluate temporal trends of various organochlorine contaminants within Lake Erie fish communities, the results of which generally indicated decreasing trends through time. The second half of this study used a similar Bayesian approach to propose a framework for updating fish consumption advisories, with specific attention paid to the acknowledgment of uncertainty and natural variability when producing such consumption guidelines. |
