The effects of tides on vertical motion in the Mediterranean Sea

Tidal forcing in numerical models is necessary to correctly forecast the ocean, since tides are a significant source of energy and driver of mixing in the global ocean. In the Mediterranean Sea, the amplitude of tides is lower than in many other regions of the global ocean, with the exception of the Alboran Sea, North Adriatic, and Gulf of Gabes, but the effects of tides are not limited to parts of the basin with high tidal amplitude. Analysis of the relationship between tidal forcing and measures of vertical motion including mixed layer depth, vertical velocity, and vertical diffusivity have not previously been carried out throughout the Mediterranean Sea.

This work investigates the effects of tides on vertical motion in the Mediterranean Sea, using the hydrodynamic model corresponding to the Copernicus Monitoring Environment Marine Service (CMEMS) system, a baroclinic forecasting model for the Mediterranean Sea, integrated over 5 years. Several regions were selected for separate study, based on their tidal amplitude and the local importance of vertical dynamics, such as deep water formation. The inclusion of tides increased the mean mixed layer depth in winter in the Mediterranean Sea, as well as in most of the selected regions. The magnitude of vertical velocity was also increased by tides on a basin level, but did not increase consistently throughout the selected regions. Vertical diffusivity, Brunt-Väisälä frequency, and Richardson number results were also explored.

The relationship between tidal amplitude and vertical velocity was additionally studied from a theoretical perspective. This work developed a prognostic equation for vertical velocity which was then compared to model results.

Our improved understanding of the effects of tidal forcing on vertical motion in the Mediterranean Sea highlights the necessity of including tides in high resolution ocean models, and allows for the separation of their effects from the impact on vertical mixing due to other modelling choices such as the chosen parametrization of vertical diffusivity.