FORECASTING PCB CONCENTRATIONS IN LAKE MICHIGAN SALMONIDS: A DYNAMIC LINEAR MODEL APPROACH

Ecological time-series data often consist of highly variable observations collected from ecosystems whose structure and function are changing, making trend assessment and forecasting difficult. Dynamic linear models (DLMs) were used to study time trends in annual average PCB concentrations in five species of Lake Michigan salmonids, using data collected from 1972 to 1994 by both the Michigan and Wisconsin Departments of Natural Resources. DLMs use an adaptive fitting procedure to track trends in the rate of decline over time, in contrast to other approaches that fit fixed parameters. We used this method to make forecasts of PCB concentrations in these fishes, along with Bayesian credible-interval estimates associated with these forecasts. Point estimates of PCB concentrations in all five species show a decline through a 10-yr forecast horizon. However, upper bounds on the 90% credible-interval uncertainty estimates for chinook salmon and lake trout indicate that steady or slightly increasing annual averag...