Assessment of the South African chokka squid Loligo vulgaris reynaudii - Is disturbance of aggregations by the recent jig fishery having a negative impact on recruitment?

An assessment of the South African chokka squid Loligo vulgaris reynaudii f ishery is undertaken. Two fisheries catch chokka: the directed jig End the trawl fishery. Chokka is a by-catch in the latter fishery, which targets pr imarily Cape hake (Merluccius spp.) and Agulhas sole (Austroglossus pectora lis). A population dynamics model is fitted to the jig and trawl CPUE indic es, and two biomass indices from scientific surveys, by means of an observa tion-error estimator. The dynamics are modelled for two periods: January-Ma rch and April-December. Parameters estimated are the annual recruitment R a nd the catchability coefficients corresponding to each biomass index; a com posite somatic growth and mortality parameter, g, is fixed externally. With in this approach, two models for annual recruitment are considered: (A) rec ruitment is constant above a biomass threshold, below which it starts to de cline; (B) recruitment depends on jig-induced fishing mortality, larger val ues of which have an increasing negative impact on reproductive success. Th e parameters estimated for model A have wide confidence intervals and the m odel is unable to fit the decline in the early years of the trawl CPUE time -series. A better fit is achieved for the approach taken in model B. The li nkage in this model of declines in recruitment to disturbance caused by the jig fishery is, however, somewhat ad hoc, and a basis for an independent t est of this link is suggested. Model B estimates the current biomass to be heavily depleted. Stochastic projections under the assumption that the curr ent effort level is maintained, show that the risk of the spawning biomass falling below 20% of its pristine level over the next 10 years is close to 90%. Effort needs to be cut to 2/3 of that at present to achieve a substant ial reduction in this level of risk. Some alternative scenarios that are in vestigated, such as that of a non-linear relationship between jig CPUE and biomass, give more pessimistic results. (C) 2000 Elsevier Science B.V. All rights reserved.