The Uncertain Environment of Fishery Scientists and Managers

Traditionally, single-species deterministic models have been used to describe the interaction between fish populations and fish harvesters. Management theory and practice often assume that decisions are made based on exact knowledge of the fishery system (i.e., population size, catch, fishing effort, recruitment, and so on, predicated or measured without error). Unfortunately, traditional approaches of fishery science and management ignore a major aspect of fishery systems—uncertainty. Biological productivity of most fishery resources is primarily dependent on recruitment. Recruitment varies by an order of magnitude or more, with little apparent relationship to spawning stock size except at extremely reduced levels. For practical purposes, production of fish populations is stochastic, not deterministic. Furthermore, growth and natural mortality parameters, which are usually assumed constant when analyzing individual populations, actually vary as a result of population interactions. In addition, the contagious and dynamic spatial distribution of most fish populations contributes still more uncertainty. Some of the variability in recruitment can be accounted for by prerecruit surveys and correlations with environmental factors. Multispecies models which attempt to account for the prédation effect on natural mortality are now available. There are robust fishery management methods which are less dependent on precise and real time estimation of population size.