Vertical structure, turbulent mixing and fluxes during Lagrangian observations of an upwelling filament system off Northwest Iberia

In August 1998, a recurrent filament located near 42 degreesN off Galicia was sampled as part of the OMEX-Il project. Lagrangian and other observations were made on the shelf where the filament arose and offshore in the filament itself under upwelling favourable but fluctuating winds. The shelf drift experiment monitored a change from southward to weak northward net flow as the winds decreased to zero. Shipborne ADCP measurements showed that the shelf was supplying decreasing volumes of water to the filament as the wind speeds decreased. At the shelf edge the internal tide was larger than can be explained by local forcing and there were many unusually large high frequency internal waves with a quasi-sinusoidal form. Turbulence observations revealed enhanced dissipation rates and vertical eddy diffusion coefficients within the shelf thermocline (of order 1 cm(2) s(-1)), which appeared to be caused by the breaking of internal wave. A second Lagrangian experiment was executed in the filament some 120 km offshore, which again coincided with a period of wind relaxation. Cross-sections revealed a double cold core and that the offshore flow was limited to a thin surface layer. Substantial onshore flow occurred below 50 m in the centre of the filament, while the strongest and deepest -offshore flow coincided with its northern boundary. Turbulent kinetic energy dissipation rate measurements showed very weak mixing below 15 in in the filament core, but enhanced mixing at its boundaries. Four mixed layer drifters released in the filament initially indicated convergence at its southern boundary, marked by strong temperature and salinity contrasts. After the wind became more favourable for upwelling, the drifters accelerated. One drifter traced the full extent of the filament, while the other three escaped from it and began to circulate cyclonically over 28 days in a 100 km diameter loop back towards their release point. Although strong mesoscale activity linked the shelf and ocean regimes, offshore transport in the filament was weak at the time of the experiment and vertical and horizontal re-circulations on a variety of time scales were important. There was sufficient vertical mixing in the thermocline to cause it to thicken and draw some heat into the lower layers during the summer months on the shelf. The amount of heat involved was too little to have a significant impact on the development of a filament over a typical lifetime of a week. (C) 2001 Elsevier Science Ltd. All rights reserved.