Virtual population analysis requires information on the fishing mortality rate (or
abundance) for one age group from each cohort. In many cases available data are
insufficient to estimate these rates for every age group and structural assumptions must
be imposed to reduce the number of effective parameters. Past assessments of western
Atlantic bluefin tuna (Thunnus thynnus) have reduced the number of
parameters by assuming pre-specified values for the ratios of the fishing mortality rates
on adjacent age groups. More recent bluefin tuna assessments have estimated terminal
fishing mortality rates (F) on each age with a constraint that restricts
change from one year to the next. We explore the implications of these methods of
estimating terminal-year F through retrospective analyses of the 2006
bluefin tuna assessment and through stochastic simulations. The use of pre-specified
ratios for F created strong retrospective biases and may have led to
overly optimistic projections. Constraining annual changes in the terminal-year
F appeared to mute retrospective patterns and resulted in abundance
projections less prone to spurious initial leaps. Simulation results indicate that the
constraint improves estimation, particularly with moderate to low interannual changes in
selectivity.