No abstract

No abstract

Abstract. This work describes the design and validation of a high-resolution (1/36 • ) ocean forecasting model over the "Iberian-Biscay-Irish" (IBI) area. The system has been setup using the NEMO model (Nucleus for European Modelling of the Ocean). New developments have been incorporated in NEMO to make it suitable to open-as well as coastal-ocean modelling. In this paper, we pursue three main objectives: (1) to give an overview of the model configuration used for the simulations; (2) to give a broad-brush account of one particular aspect of this work, namely consistency verification; this type of validation is conducted upstream of the implementation of the system before it is used for production and routinely validated; it is meant to guide model development in identifying gross deficiencies in the modelling of several key physical processes; and (3) to show that such a regional modelling system has potential as a complement to patchy observations (an integrated approach) to give information on non-observed physical quantities and to provide links between observations by identifying broader-scale patterns and processes. We concentrate on the year 2008. We first provide domain-wide consistency verification results in terms of barotropic tides, transports, sea surface temperature and stratification. We then focus on two dynamical subregions: the Celtic shelves and the Bay of Biscay slope and deep regions. The model-data consistency is checked for variables and processes such as tidal currents, tidal fronts, internal tides and residual elevation. We also examine the representation in the model of a seasonal pattern of the Bay of Biscay circulation: the warm extension of the Iberian Poleward Current along the northern Spanish coast (Navidad event) in the winter of 2007-2008.

Abstract. Through two numerical experiments, a 1-D vertical model called NEMO1D was used to investigate physical and numerical turbulent-mixing behaviour. The results show that all the turbulent closures tested (k + l from Blanke and Delecluse, 1993, and two equation models: generic length scale closures from Umlauf and Burchard, 2003) are able to correctly reproduce the classical test of Kato and Phillips (1969) under favourable numerical conditions while some solutions may diverge depending on the degradation of the spatial and time discretization. The performances of turbulence models were then compared with data measured over a 1-year period (mid-2010 to mid-2011) at the PAPA station, located in the North Pacific Ocean. The modelled temperature and salinity were in good agreement with the observations, with a maximum temperature error between −2 and 2 • C during the stratified period (June to October). However, the results also depend on the numerical conditions. The vertical RMSE varied, for different turbulent closures, from 0.1 to 0.3 • C during the stratified period and from 0.03 to 0.15 • C during the homogeneous period. This 1-D configuration at the PAPA station (called PAPA1D) is now available in NEMO as a reference configuration including the input files and atmospheric forcing set described in this paper. Thus, all the results described can be recovered by downloading and launching PAPA1D. The configuration is described on the NEMO site (http://www.nemo-ocean.eu/Using-NEMO/ Configurations/C1D_PAPA). This package is a good starting point for further investigation of vertical processes.

A high resolution simulation covering the Iberia–Biscay–Ireland (IBI) region is set-up over July 2007–February 2009. The NEMO model is used with a 1/36&deg; horizontal resolution on 50 <i>z</i>-levels in the vertical. It is forced by the astronomical potential and atmospheric forcing fields which consist of 3-hourly ECMWF analyses. Initial hydrographic conditions are derived from an Atlantic and Mediterranean Sea analyse at 1/12&deg; from Mercator Ocean (PSY2V3 model). At the open boundaries, IBI is forced with PSY2V3 temperature and salinity fields. It is also forced with tidal currents and elevations and inverse barometer elevations. In this study we evaluate the realism of the simulation through comparisons with an extensive observational dataset including climatology, temperature and salinity profiles, satellite SST data, sea surface buoys, tide gauges, altimeter data and HF radar data. A specific interest is given to the procedure used for the validation. General aspects of the simulation and its quality are analysed and particular attention is given to the validation of high frequency processes including the diurnal cycle, barotropic and internal tides, and surges. Finally, we focus on specific aspects of the circulation on the European sea shelves and give a qualitative assessment by studying tidal fronts, and specially the Ushant front, and the winter extension of the Iberian Poleward Current along the Northern Spanish coast during winter 2007–2008