What controls the spatial distribution of spawning anchovy in the Bay of Biscay? : a multi-model approach

Abstract

Spawning is an important phase for short-lived species like anchovy (Engraulis encrasicolus) as it ensures the renewal of the population through the recruitment process. Data from the PELGAS (PELagiques GAScogne) surveys, taking place since 2000 in the Bay of Biscay, were used to identify the controls of anchovy spawning distribution. Six ecological concepts underlying fish distribution were taken into account. Two are external to the population - environment and site attachment - and four are internal to the population: spatial dependency, population size, demographic structure, and population memory. Each of these ecological hypotheses was represented by one or several explanatory variables that were related to both presence–absence and positive abundance of anchovy using generalized additive models. Models were evaluated into a multi-model approach framework using a distance measure between model predictions and real observations. The Akaike Information Criterion and the log-likelihood were used for measuring model adjustment and prediction capacity. Models were ranked and selected according to these distances and hypotheses that were present in selected models were determined as relevant to explaining spawning anchovy distribution. Presence–absence was mainly determined by physiologically constraining environmental conditions such as temperature and large-scale spatial dependency, whereas inside the areas of presence, positive abundances were determined by trophic conditions, suitable for the successive maturations of the gonads, and population structure such as age‐segregation patterns. The results of this study may show that homing could be a key mechanism that ensures a successful return of the adults to the same spawning areas year after year.