Evaluating projection methodology for Northwest Atlantic groundfish with consideration of biological assumptions and environmental influence

No abstracts are to be cited without prior reference to the author.

Population projections for several groundfish species in the Northwest Atlantic Ocean were evaluated for accuracy, and sources of bias were determined. Forecasts were made ‘retrospectively,’ i.e., assessment data were truncated and forecasts were made during a period with observed catches, observed biological factors (size, weight, maturity), and observed environmental conditions. Both the assessment model and the forecast model were age structured. This framework allowed evaluation of different forecast techniques, and the accuracy of different assumptions about biological factors and environmental interaction could be compared. A secondary consideration was the identification of a feasible time horizon for projections (how many years before forecasts become unreliable) and the implications this has for defining appropriate reference points.