Exploring the thermal cycle of the Northern North Sea area using a 3-D circulation model: the example of PROVESS NNS station

Abstract

Within the framework of activities of the EC funded Project ‘Processes of Vertical Exchange in Shelf Seas’ (PROVESS), the seasonal thermal behaviour during 1998 at a station in the North Sea has been investigated using COHERENS, a three-dimensional fully non-linear hydrodynamic model. Extensive hydrographic measurements were carried out at the Northern North Sea (NNS) station, located at (59°20′N, 1°E). The collected data are used to validate the model results, showing an acceptable agreement between modelled temperatures and those obtained from CTDs and moored thermistors. This is valid both for surface and bottom temperatures, while the mixed layer thickness appears to be underestimated. A series of 3-D runs, testing different turbulence schemes, an internal wave mixing (IWM) parameterisation and the sensitivity to an increase of the surface stress, have been performed with the aim of assessing the relative importance of the advective and mixing processes. The model comparisons mostly evidenced differences in the behaviour of the bottom layer temperature during the last part of the year, which may be due to advection processes. The adoption of an internal wave mixing parameterisation, though managing to reproduce a deeper thermocline, overestimates the mixing around the period of the thermocline breakdown. The run adopting a wind stress increased by 50% provides a better agreement between observed and modelled thermocline. This applies also for surface velocities when compared to Acoustic Doppler Current Profiler (ADCP) measurements, while the bottom ones appear slightly underestimated only in the U-component. The Northern North Sea site appears to be located close to a gyre induced by thermal fronts. Comparisons with nearby wind data measured by an oil rig and by the ships operating in the area seem to confirm that the wind forcing values adopted during the integration are underestimated.