The Mediterranean Sea shows a peculiar anomaly in its nutrient pattern compared to the global ocean, as there is decrease in nutrient concentration from west to east. This feature has been attributed to the antiestuarine circulation at the Strait of Gibraltar, where an eastward flow of Atlantic nutrient‐poor surface waters is compensated by a westward countercurrent of Mediterranean nutrient‐rich deep waters. This water exchange has been suggested as the ultimate cause for the oligotrophy of the Mediterranean basin, even though only a few studies have accurately examined the magnitude of the nutrient flux through the Strait of Gibraltar. In this work, data from the Gibraltar Fixed Time series (GIFT) between 2005 and 2008 were used to assess nutrient distributions. Applying a two‐layer model of water mass exchange and using the Mediterranean outflow recorded in situ, the net export of nutrients from the Mediterranean to the Atlantic was calculated as 139 and 4.8 Gmol yr−1 of nitrate and phosphate, respectively. The results also demonstrated that the Atlantic inflow is not nutrient depleted and in particular contains significant levels of phosphate, which is the limiting factor for biological productivity in the eastern Mediterranean. The distribution of the quasi‐conservative parameter N* in the western and eastern basins indicated that nitrate‐deficient surface waters are transformed into phosphate‐deficient bottom waters by internal cycling processes. Therefore, phosphate depletion in the Mediterranean does not have its origin in the entry of a phosphorus‐impoverished Atlantic inflow through the Strait of Gibraltar.

Abstract. The exchange of both anthropogenic and natural inorganic carbon between the Atlantic Ocean and the Mediterranean Sea through Strait of Gibraltar was studied for a period of two years under the frame of the CARBOOCEAN project. A comprehensive sampling program was conducted, which was design to collect samples at eight fixed stations located in the Strait in successive cruises periodically distributed through the year in order to ensure a good spatial and temporal coverage. As a result of this monitoring, a time series namely GIFT (GIbraltar Fixed Time series) has been established, allowing the generation of an extensive data set of the carbon system parameters in the area. Data acquired during the development of nine campaigns were analyzed in this work. Total inorganic carbon concentration (C T ) was calculated from alkalinity-pH T pairs and appropriate thermodynamic relationships, with the concentration of anthropogenic carbon (C ANT ) being also computed using two methods, the C* and the TrOCA approach. Applying a two-layer model of water mass exchange through the Strait and using a value of −0.85 Sv for the average transport of the outflowing Mediterranean water recorded in situ during the considered period, a net export of inorganic carbon from the Mediterranean Sea to the Atlantic was obtained, which amounted to 25±0.6 Tg C yr −1 . A net alkalinity output of 16±0.6 Tg C yr −1 was also observed to occur through the Strait. In contrast, the Atlantic water was found to contain a higher concentration of anthropogenic carbon than the Mediterranean water, resulting in a net flux of C ANT towards Correspondence to: I. E. Huertas

L'écosystème pélagique de la côte atlantique marocaine est influencé par la variabilité spatio-temporelle de l'upwelling. L'évolution des paramètres physico-chimiques et biologiques ainsi que leur corrélation et leur regroupement par l'analyse en composante principale nous ont permis de subdiviser le littoral atlantique en quatre zones de remontées : deux zones situées au nord de Cap Juby (28 • N), caractérisées par une activité estivale et deux zones situées au sud, actives en permanence avec une intensité variable.

Fisheries constitute an important economic sector for Morocco, where the species Sardina pilchardus represents the main landings. In acoustic evaluations conducted along the Moroccan coast since 1995, the absence of juveniles in 1996 and 1998 and the collapse of the sardine stock between 1996 and 1997 represent the main events until 2002. Sardines are known to be microphageous planktivores and thus are sensitive to environmental variability. A biogeochemical model coupled to a hydrodynamic model (ROMS) was run over the Canary Current System (1991)(1992)(1993)(1994)(1995)(1996)(1997)(1998)(1999)(2000)(2001)(2002) to investigate the environmental factors that could have played a role in the variability of the sardine spawning. A grid refinement (1 ⁄ 12°) centred on the Saharan Bank (SB) region was built to study the main spawning ground of sardines off northwest Africa. The volume of the potential spawning habitat (PSH) of sardines was defined as a function of depth, temperature and salinity, which are included in the ranges 0-200 m, 15-21°C and 35.8-36.8, respectively. Our modelling frame was able to reproduce the seasonal cycle of temperature, phytoplankton concentration and PSH over the SB. It also captured the warming associated with the negative index of the North Atlantic Oscillation of 1995-1997 and allowed a description of the inter-annual variability of the PSH. Our experiment shows that the volume of PSH was much reduced in 1996 compared to the other years. The results suggest that the delay between high abundance of plankton and favourable spawning conditions is a good proxy for determining the recruitment failure of sardine in this oceanic region.

Seawater pH is undergoing a decreasing trend due to the absorption of atmospheric CO2, a phenomenon known as ocean acidification (OA). Biogeochemical processes occurring naturally in the ocean also change pH and hence, for an accurate assessment of OA, the contribution of the natural component to the total pH variation must be quantified. In this work, we used 11 years (2005–2015) of biogeochemical measurements collected at the Strait of Gibraltar to estimate decadal trends of pH in two major Mediterranean water masses, the Western Mediterranean Deep Water (WMDW) and the Levantine Intermediate Water (LIW) and assess the magnitude of natural and anthropogenic components on the total pH change. The assessment was also performed in the North Atlantic Central Water (NACW) feeding the Mediterranean Sea. Our analysis revealed a significant human impact on all water masses in terms of accumulation of anthropogenic CO2. However, the decadal pH decline found in the WMDW and the NACW was markedly affected by natural processes, which accounted for by nearly 60% and 40% of the total pH decrease, respectively. The LIW did not exhibit a significant pH temporal trend although data indicated natural and anthropogenic perturbations on its biogeochemical signatures.