Abstract

Two indirect stock assessment models are implemented in order to analyze changes in the common sardine abundance. One model uses a calendar-year configuration (January-December year t) and the other a biological-year configuration (July year t - June year t+1). Each model is contrasted with catch-at-length data, total catch in weight, catch per unit effort from 1990 to 2002, and biomass from acoustic surveys done every January from 2000 to 2002 off central-southern Chile. The two applied population dynamics models (calendar and biological year approaches) differ structurally only in the treatment of the data that reveal the annual changes. The changes observed in the common sardine abundance, per calendar or biological year, indicate similar trends as those reported in earlier studies. Nevertheless, differences were registered in the estimations for both assessments, with the biological year showing lower abundance levels that are closely related to recruitment; this, considers the magnitude of the real composition of the annual class/cohort. On the other hand, the calendar year assessment incorporates the remaining annual classes/cohorts, thereby generating an erroneous estimation of the annual class/cohort abundance. The two models are compared with a likelihood indicator, finding that the assessment using the biological configuration was the best model, with a better fit to the data and greater degrees of explanation.