Linking the movement of South African sardine and anchovy to environmental variables using a model of intermediate complexity

Abstract

This study investigates the movement of sardine and anchovy between areas around South Africa, and its possible relationship with key environmental drivers using a Model of Intermediate Complexity for Ecosystems (MICE). This model includes biomass-based, age- and area-structured population dynamics for sardine and anchovy to account for these species' major life history migrations and links between environmental variables and fish movement. In addition to movement, the model accounts for growth, recruitment, natural and fishing mortality. The environment-linked models use sea surface temperature (SST) and an upwelling index for the southern Benguela and Agulhas Bank as drivers. SST and upwelling were selected because these have consistently been identified as potential drivers of the movement of anchovy and sardine in South Africa. Eight models were developed in this study, and models with and without links to environmental variables are compared. The model parameters were estimated by fitting to the hydro-acoustic survey recruit and total biomass data, daily egg production method estimates of anchovy spawning biomass, and sea surface temperature and primary production data from 1987 to 2014. These spatially disaggregated models adequately reproduced the historical anchovy and sardine recruit and adult biomass. The results found that the relationships estimated between anchovy movement and environmental variables were relatively stronger than those for sardine, especially for upwelling. However, none of the environment-linked models, which estimated a greater variability in the movement residuals substantially improved the representation of the dynamics of these species in the southern Benguela and the fit to the survey indices of abundance. These MICE are intended for strategic purposes and inclusion of a link between the sardine and anchovy dynamics and environmental indices could be used as the basis for evaluating the impact of future climate changes on these populations under alternative fishing scenarios.