Design of an array of profiling floats in the North Atlantic from model simulations

This study aims to analyze the contribution of different profiling float arrays to the description of the 3-D large-scale thermolialine fields. It uses outputs and profiling float simulations derived from a primitive equation model of the North Atlantic Ocean. An optimal interpolation method is used to reconstruct the large-scale and low-frequency variability of the temperature and salinity fields at different depths from various sets of simulated temperature and salinity profiles. Both the covariance function and noise-to-signal ratio are derived from the analysis of the model fields. The differences between the reconstructed fields and the reference (model) fields are then analyzed. As expected, the results are sensitive to the a priori definition of the large-scale and low-frequency reference field. They show, however, that a 3degrees array with a 10-day cycle (Argo "nominal" resolution) can retrieve most of the variance of the large-scale and low-frequency temperature and salinity signals as observed by a 1/3degrees primitive equation model. The comparison between Eulcrian and Lagrangian arrays shows only a slight deterioration in the results due to the spatial dispersion of the floats and due to the loss of 20 of them which left the model domain during the experiment. (C) 2002 Elsevier Science B.V. All rights reserved.