THE PERFORMANCE OF A PRODUCTION-MODEL MANAGEMENT PROCEDURE

The performance of a production-model management procedure (the combination of a Schaefer surplus-production function observation-error estimator and the f(0.1) harvesting strategy) is evaluated by means of Monte Carlo simulation. Several factors which impact the performance of a management procedure (initial depletion, historic catch trajectory, extent of recruitment variability, level of noise about the abundance index, and values for other population model parameters) are considered. The best performance, in terms of conservation and moving the population to its target biomass level, is achieved when recruitment variability, the noise about the abundance index and natural mortality are low, and productivity is high. Performance, in terms of conservation, is poor if management commences from a virtually unexploited state due to the possibility that the estimator overestimates the size of the biomass. In this situation, performance is improved by restricting the extent of inter-annual change in catch, although the improvement is at the expense of lesser recovery of overexploited resources and lesser utilization of resources which are initially virtually pristine. The results of this paper can be used as part of an evaluation of whether a production model can be used in the management of a particular resource.