The Multispecies Fisheries Problem: A Case Study of Georges Bank

Abstract

A multispecies data base is the foundation of the Georges Bank fisheries research. Fisheries statistics describe the amount of each species caught, location of capture and fishing effort for most fishing trips to Georges Bank. Samples are taken from the catch in order to estimate its size and age composition.Standardized research vessel bottom trawl surveys of the Georges Bank region have been conducted, at least annually, since autumn 1963. A stratified random sampling design is applied. The coefficient of variation of the sample mean is about 30% for the primary demersal species (e.g., haddock, cod, yellowtail flounder).During bottom trawl surveys, fish stomachs are collected. Over the last two decades, the content of tens of thousands of stomachs of about 80 species has been determined quantitatively.Technological interactions (simultaneous capture of multiple species) prohibit achieving target catches, based on single species conservation criteria, of all of the individual species of Georges Bank. Linear programming was used to determine the “optimum” combination of catches that would not violate catch constraints of any individual species. Cluster analysis was used to define assemblages of species which are treated as a multispecies fishery unit.Feeding habits data indicate the potential for biological interactions. The estimate of MSY for the aggregate of the finfish and squid of the region is lower than the sum of the estimates of MSY for individual species, possibly as a result of biological interactions. Yet an empirical approach fails to demonstrate, explicitly, biological interactions. Nevertheless, the energy budget of Georges Bank indicates that predation by a few species of fish are a major cause of the mortality of prerecruits.KeywordsBiological InteractionFishing EffortTrawl SurveyOtter TrawlFishery StatisticThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.