The emphasis of this year's meeting was on sampling intensity and sampling methods. Sampling schemes in the various fisheries were reviewed, and a number of serious deficiencies were identified. A list of recommendations was drafted, including specifications for minimum sampling requirements. The working group also decided to standardized length measurements in the region, and to adopt the system of measuring total length to 1 cm down (0.5 cm down for sardine and anchovy) starting from 1 January 2004. The measures recommended by the working group are aimed at bringing sampling to an adequate level for fish stock assessment in all fisheries covered by the working group.
The working group still had problems in applying age structured stock assessment methods to the fish stocks in the region. This may be due to problems in age determination of the fish. As long as age determination of the fish remains uncertain, not much confidence can be placed in the outcome of age-structured models. Some progress has already been made in the age determination of sardine, thanks to the results produced by the age reading workshop in Kaliningrad in 2001. For sardinella an age reading workshop was held in Dakar in February 2003 and the results of this workshop look promising. It should be realized, however, that once a reliable age reading method has been established, it will take some years before adequate time series can be constructed.
For sardine, sardinella and horse mackerel, inconsistencies in the catch at age matrices were very large, and the working group decided that the data were too unreliable to use them in an age-structured model. In the absence of better methods, the working group decided to use production models for sardinella and horse mackerel. It was realized, however, that these models might not be very appropriate to estimate MSY or optimum fishing effort. The reasons for this are well known: the models are based on the assumption of constant recruitment, and they require data series that cover both periods of high and low fishing effort. Both conditions are not fulfilled for the stocks in the West African area. The weakness of the models is shown by the widely varying results that are produced by time series that cover different periods or by different assumptions on input parameters.
For the near future, emphasis should be placed on the study of age and growth in the various species in order to obtain better data for age-structured models. In addition to the study of otoliths, research should be devoted to following length cohorts in acoustic surveys and in the commercial fishery. The increased frequency of the R/V DR. FRIDTJOF NANSEN surveys and various national surveys offers the opportunity to follow cohorts of (juvenile) fish throughout the year. This work may provide information not only on growth rate of the fish, but also on migration and total mortality. If cohorts of juvenile fish can be detected and quantified at an early stage, it may be possible to make short-term stock predictions and to advise on annual quota for the next years. Following cohorts of fish from one area to another will also provide more insight on the exchange of fish between neighbouring areas, and thereby on stock identity.
The overall conclusion of the 2003 working group meeting was that prospects for increasing our knowledge about the biology of the various species are promising. Much progress is expected from a detailed analysis of data provided by the increased acoustic surveys. The increased biological data will allow the application of age-structured models, and thereby increase the accuracy of the assessments.
