Compared with other interesting parts of the World Ocean, little is known of the eastern Mediterranean and major issues of the Mediterranean circulation are still unsolved. Among them, the most crucial one is: what is the dominant driving mechanism of the eastern Mediterranean general circulation: (1) the wind stress; (2) the thermohaline surface fluxes; (3) the inflow forcing at the Sicily Straits? What is the relative importance of these three forcing functions? Is it the same in the different sub-basins comprising the eastern Mediterranean? What modelling factors are important for the simulation of the seasonal cycle and is the general circulation overall dominated by the annual mean or seasonal signal? To answer the above questions we have carried out an extensive and thorough series of numerical experiments using a multilevel model of the circulation, suitable for coarse-resolution studies but endowed with active thermodynamics and allowing for realistic geometry (coastlines, islands, bottom relief). The model is used in a three-level version as the minimum one capable of simulating the vertical superposition of different water masses observed in the eastern Mediterranean. The climatological monthly averages of wind-stress, thermal and evaporative fluxes and inflow at Sicily are used to drive the model. In Part I of the present study it was shown that the seasonal cycle present in the wind-stress curl induces a strongly seasonal barotropic circulation comprising the entire eastern Mediterranean. This seasonal gyre reverses from being cyclonic in winter to anticyclonic in summer. The inclusion of baroclinicity, however, profoundly modifies the purely wind-driven, barotropic circulation, eliminating the strong seasonality and the winter-to-summer reversal. The first important result is that the general circulation pattern now consists of a succession of sub-basin-scale gyres, with a seasonal modulation emphasizing the cyclonic centres in winter and the anticyclonic ones in summer, according to the vorticity input by the wind. When surface thermal forcing is included, the winter-to-summer differences become very small and the yearly pattern is dominant. The second important result is that the intercomparisons of the various numerical experiments in which each driving mechanism is studied in isolation from the others allows us to classify unambiguously the three forcing mechanisms in order of relative importance in driving the circulation in the different sub-basins of the eastern Mediterranean. Specifically, for the Ionian Sea and Sea of Crete the dominant forcing is the inflow at the Straits of Sicily while for the Levantine Sea thermohaline fluxes are the main driving function. The wind-stress forcing induces a seasonal fluctuation only in the meandering path of the Atlantic jet entering the Ionian Sea through the Sicily Straits. We finally carry out a ‘central experiment’, the most realistic one in which all forcing functions drive the circulation that we compare quantitatively with other model results and qualitatively with observations. Major features can be recognized and are shown to be persistent all year long. These features are also found in the dynamic heights of the general hydrographic surveys of the Physical Oceanography of the Eastern Mediterranean (POEM) programme. The only POEM feature not reproduced by the model, an intense anticyclonic region in the south eastern Levantine, may be attributed to errors and specifically underestimates, of the available thermal fluxes whose effect is partially overcome by the wind-stress forcing. This anticyclonic cell is in fact obtained when the model is driven by the thermal fluxes alone. Overall, the model results compare well with the observational evidence provided by the POEM surveys and the thermohaline vertical circulation cell reproduced by the model is consistent with the preliminary results of the transient-tracer survey POEM-V-87. Finally, many of the persistent features found in the model circulation patterns can be related to the strong control exerted by the ambient potential vorticity f/ H upon the eastern Mediterranean circulation.
Read full abstract