Douglas S. Butterworth

Stock assessment and risk analysis for the South Atlantic population of albacore Thunnus alalunga using an age-structured production model

The South Atlantic population of albacore is assessed by means of an (observation-error) age-structured production model. Relative abundance (cpue) series up to 1993 from the Republic of China, Japan and South Africa are used in the model-fitting process. Coefficients of variation and confidence intervals for the estimates are obtained using the (conditioned) parametric bootstrap and the percentile methods respectively. The sensitivity of the results to violation of some of the assumptions made by the model, changes to the values of its input parameters, and changes to the abundance series used in the model-fitting process are examined. The results of the age-structured production model are contrasted with those of an age-aggregated observation-error estimator and those of a Bayesian estimator which allows also for fluctuations in recruitment. The resource was estimated to be markedly depleted — to probably slightly above 20% of its pre-exploitation level expressed in terms of exploitable biomass — and to...

Competition with Fisheries

Publisher Summary It is currently virtually impossible to wholly substantiate claims that predation by marine mammals is adversely impacting a fishery or vice versa . Despite a persistent notion worldwide that there is a mass-for-mass equivalence in the prey of marine mammals and the yields available to fishers, the evidence points to much more complicated situations in which this is hardly likely to be the case. Furthermore, the complexity of ecosystems could well be such that the response to a marine mammal cull, e.g., could be highly diffused through the food web, involving many other species. In some cases, competition effects are reduced because, e.g., one of the putative competitors in fact reduces the abundance of a predatory fish species, in turn affecting the abundance of the target prey species. It is worth noting that although marine mammals are the most obvious scapegoat of fishers because of their visibility, there is typically greater competitive overlap in the feeding “niches” of fish predators and fishermen. As more and better information on marine mammal diets becomes increasingly available, one of the key uncertainties in resolving questions as to the degree of competitive overlap between marine mammals and fisheries relates to limited understanding at present of the feeding strategies of marine mammals.

Antarctic marine ecosystem management

This article describes a framework within which an initial strategy could be developed for managing commercial exploitation of marine living resources of the Southern Ocean, particularly of krill and fish, in accordance with Article II of the Convention for the Conservation of Antarctic Marine Living Resources (CCAMLR). Quantitative predictions involving multi–species models are needed to assess any indirect impacts of fish or krill exploitation, and also for management to restore depleted populations. This article recommends research to provide the knowledge necessary for the models (identifying key species, estimating their demographic status, and experimental interference), and suggests interim management action to delimit management areas, agree target levels for stock–size, and monitor stocks. Early efforts to model the fishing operation are particularly recommended.

Investigation on the effects of different levels of effort and of the closed season in the jig fishery for chokka squid loligo vulgaris reynaudii

After effort control, the closed season is the most important management tool used to regulate fishing mortality in the South African jig fishery for chokka squid Loligo vulgaris reynaudii. The dynamics of the stock biomass on the spawning grounds were modelled in order to assess the effects of current effort levels and the existing closed season on the resource. The model assumes that process error dominates over observation error. The model has three parameters: g, a parameter which combines natural mortality, emigration and somatic growth; q, the catchability; I, which represents immigration to the spawning grounds. Estimates of the parameters were obtained by fitting the model to catch per unit effort data, and were used to project the stock biomass forward in time to evaluate the impact of various effort levels both with and without a closed season. Results from the study indicate that the biological and economic gains provided by the current closed season are small and that there is little justification for urgent action to curtail current effort levels. However, in view of the apparently high risk associated with the current levels of effort and the great sensitivity of the results to basic model assumptions, both maintenance of the closed season and avoiding increases above the current level of effort are recommended. Further, the level of effort may need to be reduced, and the option of lengthening an existing closed season to effect this may well prove easier to implement than any attempt to reduce the present number of participants in the fishery.

The effect of hook size on the size-specific selectivity of hottentot Pachymetopon blochii (Val.) and on yield per recruit

A model for the relationship between hook size and length distribution of catches for a line fishery was applied to experimental data for hottentot Pachymetopon blochii. Hook size influenced the length frequency of catches. The selectivity function was modelled to an adequate approximation by a gamma distribution. Exploitation rates at various fishing centres off the Western Cape, South Africa, were investigated in the context of an observed variation in the mean mass of the individual fish caught. When allowance was made for the various hook sizes used at different locations, only the resource at Gans Bay appeared to be overexploited in yield-per-recruit terms. If hook size affects the length distribution of the catch, it is important to incorporate this aspect into assessments of resource status.