Final Report Summary - INMEDIATO (Influence of the Mediterranean Outflow on the Atlantic Ocean Climate: the role of local scale processes)
The Mediterranean outflow (MO) water spreads as a basin-scale positive salinity anomaly at about 1000-1200 m depth through the north Atlantic (NA). The basin-scale Mediterranean salinity tongue originates from a slope-bounded gravity plume in the Strait of Gibraltar and the Gulf of Cadiz. Therefore, small scale processes and local scale nonlinear interaction with tides and topography may have an impact on the MO properties and its further spreading pattern in the NA. Those processes are not accounted for in most of the global/basin scale models, where MO is usually prescribed as a source term or as a boundary condition. Therefore, the project goal was to investigate the impact of tides on the MO spreading in the NA, trying to clarify whether tidally-induced local-scale processes at Strait of Gibraltar and the Gulf of Cadiz affect the MO properties and spreading through the NA and, if so, which are the specific mechanisms involved.
The main difficulty of the project was related to the need of counting on a wide range of spatial and temporal scales, which we have approached simultaneously resolving the local-scale processes at the Strait of Gibraltar and the Gulf of Cadiz and the larger-scale processes which may influence the NA climate, by means of the global Max Planck Institute ocean model (MPI-OM) with regional high resolution around the Iberian Peninsula and ts ability to explicitly include tides.
The MPI-OM simulation of the MO is successful, in the sense that the depth and properties of the MO core are properly reproduced after the 15 year spin-up period. Later, however, while the tidal run resembles better the climatology and the established MO spreading pathways, the no tidal run presents a salinity excess at MO depths in the Gulf of Cadiz which triggers a non realistic southward salt flux. This spreading of MO waters along the African slope suppresses the fresher poleward undercurrent on the eastern boundary, feed-backing therefore the salinity increase and the southward MO spreading. This pattern is common to other global models not including tides.
A series of process-study targeted numerical experiments was designed in order to reveal the mechanisms responsible for such differences in the MO simulations with and without tides. From a detailed analysis of the experiments' high-frequency output we were able to conclude that the tidal residual currents, generated by the interaction of tides with local-scale topography, play a major role in the spreading of MO in the Gulf of Cadiz. In fact, tidal residual currents advect most of the MO west (supported by observations). This is not happening in the no tidal runs. Instead, the reduced efficiency in the evacuation of MO from the Gulf of Cadiz results in a gradual salinity increase at MO depths, which sets up the already mentioned southwestward salt flux.
Additional support to our conclusions is given by the model skills to properly reproduce different aspects of the tidal dynamics in the Gulf of Cadiz. The surface tide, tidal currents and internal wave generation and propagation have been satisfactorily compared to observations.
In all our experiments, the tidal runs show a further northward spreading of MO in comparison to the no tidal runs, which lead to the question of the possible influence of MO on NA deep water formation processes and their variability. Therefore, the results of the global OGCMs simulating the MO spreading without tidal forcing should be carefully considered, especially when used to estimate the contribution of MO to the Nordic Seas salinity. The results of our project could be useful for the next generation of high resolution climate models, as long as they will be forced to address some issues relevant at smaller scales, such as tidally-induced processes.(http://www.mpimet.mpg.de/en/science/the-ocean-in-the-earth-system/ocean-physics/gibraltar.html(odnośnik otworzy się w nowym oknie)) Contact: Alfredo Izquierdo (alfredo.izquierdo@zmaw.de),Uwe Mikolajewicz uwe.mikolajewicz@zmaw.de)
The main difficulty of the project was related to the need of counting on a wide range of spatial and temporal scales, which we have approached simultaneously resolving the local-scale processes at the Strait of Gibraltar and the Gulf of Cadiz and the larger-scale processes which may influence the NA climate, by means of the global Max Planck Institute ocean model (MPI-OM) with regional high resolution around the Iberian Peninsula and ts ability to explicitly include tides.
The MPI-OM simulation of the MO is successful, in the sense that the depth and properties of the MO core are properly reproduced after the 15 year spin-up period. Later, however, while the tidal run resembles better the climatology and the established MO spreading pathways, the no tidal run presents a salinity excess at MO depths in the Gulf of Cadiz which triggers a non realistic southward salt flux. This spreading of MO waters along the African slope suppresses the fresher poleward undercurrent on the eastern boundary, feed-backing therefore the salinity increase and the southward MO spreading. This pattern is common to other global models not including tides.
A series of process-study targeted numerical experiments was designed in order to reveal the mechanisms responsible for such differences in the MO simulations with and without tides. From a detailed analysis of the experiments' high-frequency output we were able to conclude that the tidal residual currents, generated by the interaction of tides with local-scale topography, play a major role in the spreading of MO in the Gulf of Cadiz. In fact, tidal residual currents advect most of the MO west (supported by observations). This is not happening in the no tidal runs. Instead, the reduced efficiency in the evacuation of MO from the Gulf of Cadiz results in a gradual salinity increase at MO depths, which sets up the already mentioned southwestward salt flux.
Additional support to our conclusions is given by the model skills to properly reproduce different aspects of the tidal dynamics in the Gulf of Cadiz. The surface tide, tidal currents and internal wave generation and propagation have been satisfactorily compared to observations.
In all our experiments, the tidal runs show a further northward spreading of MO in comparison to the no tidal runs, which lead to the question of the possible influence of MO on NA deep water formation processes and their variability. Therefore, the results of the global OGCMs simulating the MO spreading without tidal forcing should be carefully considered, especially when used to estimate the contribution of MO to the Nordic Seas salinity. The results of our project could be useful for the next generation of high resolution climate models, as long as they will be forced to address some issues relevant at smaller scales, such as tidally-induced processes.(http://www.mpimet.mpg.de/en/science/the-ocean-in-the-earth-system/ocean-physics/gibraltar.html(odnośnik otworzy się w nowym oknie)) Contact: Alfredo Izquierdo (alfredo.izquierdo@zmaw.de),Uwe Mikolajewicz uwe.mikolajewicz@zmaw.de)