Abstract
We examined escapement policies for a stock-recruitment model with negative between-year interactions. Regardless of the degree of interaction present, the optimal policy is to always equalize escapement. Parameter estimates obtained for the Adams River sockeye salmon (Oncorhynchus nerka) indicate that between-year interactions may occur, but confidence regions include the null hypothesis of no interaction at all (the Ricker model). We conclude that the extreme amplitude of the current recruitment cycle in this stock frustrates statistical identification of interaction. It seems unlikely that between-year interactions will be measurable until the off year runs increase by at least two to three orders of magnitude. Comparison of total yields for the Adams River sockeye shows that an equal escapement policy could increase yields by at least 35% over that obtained by the current cyclic escapement pattern. This is equivalent to obtaining an additional $27 million in total yield per annum from the Adams River stock alone and, assuming a discount rate of 4%, translates into an increase in net present value of $675 million. If between-year interactions do not exist, the potential benefits of moving to an equal escapement policy are even larger, on the order of $3 to $4 billion.