Forecasts of Recruitment in South African Anchovy from Sarp

sagax and round herring Etrumeus whiteheadi are the major components of the pelagic fishery in the southern Benguela upwelling system, which encompasses the west coast of South Africa and the western Agulhas Bank at the southernmost extremity. Commercial catches since the 1970s have, until recently, been dominated by anchovy. The species is relatively short-lived (≈ 3 years) and the fishery is largely dependent upon juvenile fish (approx. 6 months old), which are spawned during a distinct season. As a result, the fishery is seasonal and highly sensitive to interannual variations in recruitment, both of which impose a number of problems in the management of the resource. The most important management tool is the setting of the Total Allowable Catch (TAC) for each fishing season (January–September/October), which needs to be done at the start of the season (January), before recruitment strength is known. Prior to 1983, the anchovy fishery was managed by adopting a constant catch strategy, which allowed for catches of approximately 300 000 tons per year (Crawford et al. 1987). In 1983, a new Sea Fisheries (SF) vessel, the F.R.S. Africana, with hydro-acoustic capacity was commissioned, and since then hydroacoustic surveys have been conducted annually during the anchovy spawning season (in November) and during the subsequent recruitment period (in May/June, Hampton 1987, 1992). Quantitative assessment of the resources has permitted development of a management procedure for anchovy based on survey estimates of stock sizes of spawners and recruits. By the early 1990s, the management procedures incorporated a strategy of constant escapement (Butterworth and Bergh 1993, Butterworth et al. 1993). An important assumption in the setting of the TAC was that recruitment in any given year was equal to the observed median recruitment. Survey results of spawner biomass in November were therefore used to determine an initial TAC for the following season. This was revised after the midyear survey in which actual recruitment was estimated (see Butterworth et al. 1993). Simulation studies by Cochrane and Starfield (1992) indicated that a valuable increase in mean annual yield could be achieved if recruitment variability could be correctly predicted at the start of the fishing season. Conservative estimates indicated a 16% increase in mean annual catch if forecasts could differentiate between “below-median” or “median or above” recruitment. Using field data collected during November surveys from 1985 to 1992, Cochrane and Hutchings (1995) developed a conceptual model of the factors influencing anchovy recruitment, and investigated 10 biological and environmental indicators that could be used to make such forecasts. Fluctuations in anchovy recruitment were shown to be associated with fluctuations in five of the indicators, viz. copepod biomass and production on the anchovy spawning grounds, the incidence of oocyte atresia in adult female anchovy, Benguela Dynamics Pillar, S. C., Moloney, C. L., Payne, A. I. L. and F. A. Shillington (Eds). S. Afr. J. mar. Sci. 19: 245–261 1998 245