Model and method in reef fishery assessment

Summary

Mathematical models are used extensively to estimate the size and productivity of fishery stocks and to assess the potential effects of alternative management strategies. Although mathematical modelling in general is quite advanced, models employed in the management of tropical fisheries tend to be quite simple. A principal limitation is the amount and quality of data. The lack of information is determined by the spatial and biological diversity of resources, the complexity of the fishery and the limited capacity for data collection and analysis. Nevertheless, focused sampling strategies can yield data suitable for analysis and management advice. Both length-based and otolith-based methodologies are available for estimating aspects of growth, mortality and recruitment. Length-based methods are simpler, but their many assumptions may be difficult to meet, while otolith analysis is more costly and time consuming. The most common assessment approaches are based on single-species surplus-production or yield-per-recruit models, even though these might not be appropriate in a multitrophic, multispecies context. Multispecies models either require a prohibitive amount of data or they lump all species together at the cost of predictive power. Useful approaches have been identified, these problems notwithstanding, often involving aspects of both single-species assessment and small-scale lumping. Even the simplest models provide some useful management information; if nothing else, they provide a target against which the need for regulation or development may be gauged.