Abstract
Overexploitation rates in the General Fisheries Commission for the Mediterranean and Black Seas (GFCM)'s marine basins are the highest in Europe, despite the Common Fisheries Policy (CFP)'s sustainability objectives of achieving Maximum Sustainable Yield (MSY) by 2020. Demersal stocks, such as hake (Merluccius merluccius), have been managed with control measures since 2008, but the biomass of hake has not recovered and catches of juveniles have not decreased.The management plan for demersal fisheries in the north-western Mediterranean, effective from 2020, has gradually reduced trawling effort and introduced time-area closures to increase biomass and reduce catches of juveniles. However, studies on fleet dynamics and the spatial reallocation of effort are limited. Hake are widely distributed, but little is known about their spatial distribution of spawners due to the absence of older individuals in the catches.To assess the impact of management, studies of fleet dynamics and spatial reallocation are needed. This thesis developed new modelling and data methods to better understand and manage the hake fishery in the Gulf of Lion, focusing on fleet dynamics, spatio-temporal closures and effort reallocation. The measures were assessed for their ability to restore hake populations and maintain economic viability.All-at-once effort reduction was the most effective in increasing catches in the long term, but caused significant initial losses. Spatial closures did not significantly reduce catches of juveniles and resulted in an uneven distribution of losses. The Spanish fleet was more affected by effort reductions, while the French fleet was more affected by spatial closures. The combined measures were less effective than effort reduction alone, but overall improvements were observed.The current model assumes no migration between population areas and a fixed dispersion of recruits, requiring an assessment of initial distribution assumptions to define smaller population areas. Alternative methods of data aggregation were explored to describe the biology and distribution patterns of hake. Significant correlations among the sizes assessed were found, and spatio-temporal distribution patterns were modelled using Data Interpolating Empirical Orthogonal Function (DINEOF)s and Empirical Orthogonal Function (EOF)s, with recurrent structures identifying via K-means clustering.The results showed that the assumptions of homogeneous distribution and dispersion of juveniles were incorrect. Young hake were more concentrated towards the centre of the Gulf of Lion, with specific winter patterns and deeper distributions for older hake. Areas of spawner fidelity were identified off the Rhône mouth and along the west coast, consistent with studies linking distribution patterns to environmental factors.In conclusion, this thesis has identified key points to help develop sustainable management strategies for hake, with new population areas for future models.