Background

For new and lightly exploited fisheries in developing countries, little data will be available upon which fisheries managers may base their decision making. However, appropriate precautionary management is required to prevent the over-exploitation of a potentially valuable, sustainable fishery.

For sustainable management, there is therefore a need for methods to inform decisions about what targets for effort levels and yields should be adopted, and how the fishery should move towards those targets in the early stages of development. There is also a need for methods to evaluate fishery monitoring procedures and decision rules, so that both appropriate monitoring and rules that allow timely reaction to unexpected trends in stock levels can be implemented.

Approach

The project developed a Bayesian statistical methodology for the assessment of new and lightly exploited stocks. This incorporates current knowledge about key fisheries parameters, derived either from the fishery to be assessed, or from information about similar fisheries elsewhere, and is therefore appropriate for fisheries with limited data. The basis and application of the methodology was described in a number of scientific papers.

The methodology was applied to the first of two case studies, the newly discovered Namibian orange roughy fishery. The methods were used to evaluate alternative fishery development strategies, and formed the primary basis for scientific advice given to the Ministry of Fisheries and Marine Resources. A second case study, of the Tongan seamount fishery, was studied using a detailed spatial software model to investigate whether the total catch and effort data being collected was adequate, or whether data were required by seamount.

Findings

The project successfully developed a methodology for the assessment of new or lightly exploited fisheries in developing countries. Its applicability has been clearly demonstrated in the Namibian orange roughy fishery, where it has been adopted to develop advice on catch limits for the Ministry of Fisheries and Marine Resources. A finding of the application of this methodology is that effort-based management can produce significantly greater yields (up to 40% greater) than catch-based management, for the same levels of risk of stock depletion.

The project has also built capacity amongst Namibian scientists, who were trained in Bayesian assessment techniques during the project, and who are now able to perform the stock assessments independently.

For the Tongan seamount fishery, whilst there was no evidence of overexploitation of individual seamounts, the model suggested that this could change rapidly, for example driven by increases in fish prices or profitability. Overexploitation of individual seamounts would only be detected when detailed spatial data on catches are available. The software model and tutorial were delivered during interactive dissemination workshops in Tonga and Fiji, and management guidelines derived from the analysis were presented.