Articles | Volume 2, issue 2
Ocean Sci., 2, 237–248, 2006
https://doi.org/10.5194/os-2-237-2006

Special issue: Mediterranean Ocean Forecasting System: toward environmental...

Ocean Sci., 2, 237–248, 2006
https://doi.org/10.5194/os-2-237-2006

  05 Dec 2006

05 Dec 2006

Assessment of the impact of TS assimilation from ARGO floats in the Mediterranean Sea

A. Griffa1,5, A. Molcard2, F. Raicich3, and V. Rupolo4 A. Griffa et al.
  • 1CNR, ISMAR/La Spezia, Italy
  • 2LSEET, Univ. Toulon, France
  • 3CNR, ISMAR/Trieste, Italy
  • 4ENEA, Casaccia, Roma, Italy
  • 5RSMAS, University of Miami, FL, USA

Abstract. In this paper, the impact of assimilating Temperature (T) and Salinity (S) profiles from Argo floats in the Mediterranean Sea is quantitatively investigated using the Observing System Simulation Experiments (OSSE) approach. The impact of varying the number of floats and their launch positions is considered, using numerical simulations with a MOM model and a reduced-order multivariate Optimal Interpolation scheme (SOFA) for assimilation. Realistic float coverage and launch positions used during the first MFSTEP phase are considered, as well as "ideal" density coverage that can be envisioned for the future, corresponding to a double coverage with respect to MFSTEP and with floats released along the VOS tracks. The most effective float trajectories are identified, showing that frontal regions play a major role, and that it is crucial to maintain a sufficient coverage of them. In addition to this, a comparison is also performed between the results obtained from MEDARGO floats in ideal conditions and results from "ideal" profiles taken at fixed points along the VOS tracks, as for the XBT data. For consistency the coverage considered is double the actual XBT coverage during MFSTEP, resulting in a threefold increase in the number of profiles compared to the MEDARGO experiment. The maximum error reduction is of approximately 10%, suggesting that spatially coarser profiles from floats can be more efficient since they follow flow features.