Search Results

You are looking at 1 - 5 of 5 items for

  • Author or Editor: Pierre De Mey x
  • Refine by Access: All Content x
Clear All Modify Search
Michel Gavart and Pierre De Mey

Abstract

An empirical orthogonal function analysis of the vertical space–time variability of the thermohaline structure of the ocean in summer and fall 1993 in the Azores Current region has been performed. The analysis used data from CTD casts of the SEMAPHORE intensive experiment, referred to as a seasonally varying climatology. The relative merits of depth-coordinate and isopycnal-coordinate representations have been examined in regard to the quality of identification of physical processes and to the effectiveness of the extrapolation from a sea level anomaly in an attempt to improve the treatment of the vertical dimension in oceanic assimilation problems. The isopycnal EOFs consistently proved more efficient in capturing the vertical structure of both dominant processes in the area of investigation: the variability linked to the Mediterranean Water, and the coherent physical system made up of the Azores Front and the Azores Current. The isopycnal analysis also proved more robust when meddies (Mediterranean Water lenses) were included in the analysis. In addition, isopycnal EOFs appeared to be more observable from altimetry: Residual sea level anomaly variance after projection on the dominant mode (representing the Azores Front/Azores Current) was 8.4 cm2 in depth-coordinate representation and 3.6 cm2 in isopycnal-coordinate representation. As an attempt to estimate the vertical structure of errors needed by assimilation schemes, differential isopycnal EOFs were calculated from pairs of casts close in space and time. Despite that only 15 pairs were available for the chosen radii, the dominant process was a quasi-homogeneous vertical displacement of isopycnals with quasi-conservation of water masses and potential vorticity on the isopycnal grid.

Full access
Vincent Echevin, Pierre De Mey, and Geir Evensen

Abstract

The representer functions used to propagate the information from observations onto model variables in a sequential data assimilation scheme are calculated with a statistical method. The Princeton Ocean Model is used in a simple configuration simulating the North Current flowing along the French coasts in the northwestern Mediterranean Sea. The influence of sea level measurements on the models’ three-dimensional temperature and velocity fields is investigated. Inhomogeneities and directions of anisotropy are evidenced. A comparison with simplified reduced order assimilation methods suggests that such schemes will not allow for consistent assimilation of nearshore altimetric data.

Full access
Pierre De Mey and Allan R. Robinson

Abstract

Techniques for the synoptic analysis, vertical inference, dynamical adjustment, and forecast of altimetric and deeper in situ data are presented as a first step towards the design of continuous assimilation schemes in limited-area oceanic domains. A year-long time series of streamfunction maps, denoted Mark 2, drawn in the POLYMODE area of the western North Atlantic is used as a benchmark for various tests and simulations. An original projection/extension scheme using empirical modes of density and/or pressure anomalies is used to obtain a first guess of the three-dimensional structure of the currents, starting from surface topography only. In the Mark 2 domain, this technique works well, since the first empirical mode is surface intensified and largely dominant. An alternative approach is to incorporate deeper data, e.g., float trajectories below the main thermocline. The first-guess currents are specified as initial and boundary condition of the Harvard 6-level quasi-geostrophic open ocean model. When surface data only are assimilated, the model makes the deep currents converge dramatically towards Mark 2. The fastest adjustment occurs in 9 days and involves nonlinear mode-mode interactions. The sensitivity of the assimilation scheme to different dynamical regimes, to bottom topography, and to the modal assumptions of the empirical extension is investigated. It is found the results are extremely robust. The presence of bottom topography further increases the rate of convergence of the deep levels. Finally, we use a simple orbital model to generate realistic altimeter track sequences. Mark 2 is sampled, and linear optimal estimation is applied to restore the surface topography. The deep adjustment still occurs. It is also found to be rather insensitive to the choice of sampling strategy and to the horizontal correlation scale.

Full access
Jean-Pierre Malardé, Claire Perigaud, Pierre De Mey, and Jean-Francois Minster

Abstract

Synoptic maps of the mesoscale dynamic topography in a band between 7.5°N and 7.5°S in the Pacific Ocean are drawn from Seasat altimeter data. They show a set of eddies 600 km in diameter and 15–20 cm in amplitude moving westward with a velocity of about 40 km day−1 along 4.5°N. Their occurrence is consistent with the surface temperature front undulations observed by Legeckis. South of the equator the signal is less coherent, but a significant degree of symmetry with the north is evidenced. The dynamics of the wave system might also present some degree of nonlinearity, as some water seems to be carried along with the wave.

Full access
Matthieu Le Hénaff, Pierre De Mey, Baptiste Mourre, and Pierre-Yves Le Traon

Abstract

The authors investigate the potential qualitative improvement brought by wide-swath, interferometry-based ocean altimetry measurements with respect to classical nadir altimeters in a coastal/shelf data assimilation system. In addition, particular attention is paid to roll errors, which could significantly reduce the expected benefits of wide-swath altimetry. A barotropic, nonlinear free-surface model is set up over the European shelf as part of an ensemble Kalman filter. Experiments assimilating simulated data are performed over the North Sea to test the ability of altimeter configurations to reduce model errors due to the action of meteorological forcing in the presence of bathymetric uncertainties. A simplified wide-swath observation scheme is used, composed of nadir altimeter height plus a nadir-centered cross-track sea level slope measurement. The simplified wide-swath measurements are found to be able to constrain events unsampled by a single nadir altimeter owing to a wider domain of influence in the cross-track direction and the ability to detect cross-track gradients.

Since the satellite-borne interferometer is highly sensitive to the platform behavior, especially satellite roll, experiments taking roll errors into account are then carried out. Whereas observational errors are considered independent in most data assimilation studies, the roll of the platform correlates those errors along the path of the satellite. Despite the large amplitude of the roll errors, the contribution of the wide-swath altimeter in coastal zones remains valuable as long as the roll frequency is known (within Gaussian error) and the assimilation scheme is designed to take observational error correlations into account.

Full access