ASSIMILATION OF SEA SURFACE HEIGHT ANOMALIES INTO HYCOM WITH AN OPTIMAL INTERPOLATION SCHEME OVER THE ATLANTIC OCEAN METAREA V

Abstract

Along-track sea surface height anomaly (SSHA) data from the Jason-1 and Jason-2 satellites were assimilated into the ocean model HYCOM from July 1, 2009 until December 31, 2009. A new and simple approach to overcome the bias between the model and observed SSHA was proposed. It focuses on the meso-scale differences between the data and the model along each satellite track. An optimal interpolation method and the Cooper and Haines (1996) scheme (C&H) were employed to produce a SSHA analysis field and to adjust model layer thicknesses over the Atlantic METAREA V. The corrected model state was used as initial condition for the next assimilation cycle. SSHA data with a 7-day window were assimilated in 3-day intervals centered in the SSHA data window and the C&H scheme was applied taking the SSHA analysis. A control run without assimilation was also performed. The results showed that the model SSHA was completely reorganized by the end of the experiment. The modifications of SSHA were compared to the American Navy HYCOM+NCODA system and AVISO data. Maximum error was reduced from 0.7 m to 0.2 m by assimilation. Comparisons were also made with the Argo temperature and salinity vertical profiles. Improvements in the currents and volume transport were also produced by assimilation. The impact in temperature was in general positive, but there was no substantial modification in salinity.