87-19 Using Simulation Models to Investigate the Impacts of P*-Based Control Rules on Stocks of Differing Life History Strategies
A transparent approach to complying with these stipulations involves calculating the catch limit that corresponds to a specified probability, P*, of exceeding a “true” OFL located within an estimated distribution (P* = P (ABC > OFLTrue)). Thus, scientists describe the uncertainty in their estimations, but managers choose the value of P*, deciding the acceptable level of “risk.” This approach is conceptually appealing, and several regional management councils already employ P*-based control rules for determining ABCs. However, the methods for its implementation are not standardized, and the implications of specific P* choices are not well understood. Equal values of P* do not necessarily provide equal levels of precaution for the resource, nor do they imply equal impacts on fishermen and the fishery. Rather, the ramifications of P* choices depend heavily on the amount of uncertainty in the stock assessment and on the vulnerability of the species in question.
Drawing from classic fish life history theory, equilibrium strategists will have the hardest time recovering from years when the OFL is exceeded. Periodic strategists can better sustain occasional excesses but do not fare well under heavy sustained fishing pressure. Finally, opportunistic strategists are best able to withstand occasional overfishing. For this analysis, three South Atlantic species were chosen to represent these strategies: sandbar shark Carcharhinus plumbeus, vermilion snapper Rhomboplites aurorubens, and Atlantic menhaden Brevoortia tyrannus. Biological parameters drawn from recent stock assessments were incorporated into simplified statistical catch-at-age models. Reported OFL estimates were assigned a range of distributions to test degrees of uncertainty. The simulated populations were then fished according to ABCs derived from a range of P* values. In this way, predicted trends could be better quantified, and ultimately, the results can be generalized to inform management and support control rule decisions.