Résumé
European hake (Merluccius merluccius) is a key species in the management of several mixed fisheries in the north-east Atlantic where regional spatial management plans are being set up. The wide spatial distribution of hake and the lack of knowledge of some processes involved in its spatial dynamics could hamper the assessment and management of hake.To help understand the processes, we implemented a spatial version of the Stock Synthesis stock assessment model for northern hake, including available data on this stock and the associated fisheries. For the 1978–2012 time series, the model distinguished the Bay of Biscay, the Celtic Sea and a northern area aggregating the West of Scotland and the North Sea. The model considered an age-structured population for each age class and area, a seasonal migration, global recruitment allocated to areas and fishing fleets for each area. Parameters for stock and exploitation spatio-temporal dynamics were estimated by likelihood maximization for each quarter and area.The comparison of spatial and non spatial models goodness of fit showed they were close, although the fit to abundance indices slightly improved with the spatial model.Fishing mortalities, spawning stock biomasses and recruitments were slightly different from the current assessment model estimates, essentially at the beginning and the end of the time series. On average over the time series, fishing mortality estimates relative to the abundance were higher in the Bay of Biscay than elsewhere, spawning stock biomass estimates were higher in the Celtic Sea than elsewhere, and recruitment occurred mainly in the Bay of Biscay. Finally, the final spatial model estimated a strong movement from the northern area to the Celtic Sea. The final spatial model is a step towards a spatial assessment and management of the stock.