1. Abstract Combination of in-situ and satellite observations to monitor the Ocean State: Application to the North Atlantic Ocean Sandrine Mulet, Stéphanie Guinehut, Marie-Hélène Rio, Anne-Lise Dhomps, Laurent Bessieres, Gaël Nicolas and Gilles Larnicol CLS / Space Oceanography Division, Ramonville Saint-Agne, France 2 products obtained by merging different sources of observations have been developed and validated: ARMOR3D (3D T/S fields) and Surcouf3D (absolute height and current). These kind of products, complementary to numerical models, are very useful to depict evolution in the ocean state. They can provide information both on the physical variables (i.e. Temperature Variability patterns) and on ocean integrated parameters (Meridional Overturning strength). We underlined a limitation of the method in the western boundary current off the Bahamas  Understand why and try to improve it  Compute error bars Carton J.A. and B.S. Giese, 2008: A reanalysis of ocean climate using Simple Ocean Data Assimilation (SODA), Monthly Weather Review, 136, 2999-3017. Cunningham S.A., T. Kanzow, D. Rayner, M.O. Baringer, W.E. Johns, J. Marotzke, H.R. Longworth, E.M. Grant, J.J.-M Hirschi., L.M. Beal, C.S. Meinen, H.L Bryden (2007), Temporal variability of the Atlantic Meridional Overturning Circulation at 26°N. Science, 317, 935-938. Dhomps A.L., S. Guinehut, P.Y. Le Traon and G. Larnicol, 2010: A global comparison of Argo and satellite altimetry obsercations, submitted to Ocean Science. Dhomps A.L., 2010: Amélioration des méthodes de combinaison des données Argo et altimétrie pour le suivi des variations à long terme de l’océan. Thèse de l’Université Paul Sabatier, Toulouse III Ferry N., Parent L., Garric G., Barnier B., Jourdain N. C. and the Mercator Ocean team, 2010: Mercator Global Eddy Permitting Ocean Reanalysis GLORYS1V1: Description and Results. Mercator Ocean Quarterly Newsletter #36, January 2010, 15-27 Ollitraut, M. and J.-P. Rannou, 2010: ANDRO: An Argo-based deep displacement atlas. Mercator Ocean Quarterly Newsletter #37, April 2010 27-34. Rio, M. H., and P. Schaeffer (2005), The estimation of the ocean mean dynamic topography through the combination of altimetric data, in-situ measurement and GRACE geoid, Proceedings of the GOCINA international workshop, Luxembourg. Roemmich, D. and J. Gilson, 2009: The 2004-2008 mean and annual cycle of temperature, salinity, and steric height in the global ocean from the Argo Program. Progress in Oceanography, 82, 81-100. smulet@cls.fr Producing comprehensive information about the ocean has become a top priority to monitor and predict the ocean and climate change. Complementary to modeling/assimilation approaches, an observation based approach is proposed here. It relies on the combination of in-situ (temperature and salinity profiles) and remote-sensing observations (altimetry and sea surface temperature) and statistical methods. Global temperature and salinity (ARMO3D), absolute height and current fields (SURCOUF3D) are provided at a weekly period from the surface down to 1500-meter depth and for the 1993-2009 periods. Data Method  Compute field of Absolute Dynamic Topography and the corresponding geostrophic circulation  SURCOUF3D geostrophic current fields and absolute height : [0-1500m] – 1/3° grid - weekly Argo floats Validation SURCOUF3D ● SURCOUF3D (weekly, 1/3°) ▲ GLORYS = Mercator-Ocean reanalysis (weekly, 1/4°) Ferry et al., 2010 ♦ Armor3D = velocity field deduced from the thermal wind equation assuming that 1500m is a level of no-motion (weekly, 1/3°) ● ANDRO = drifting velocities from the Argo floats (≈10days, ≈50/100km) Ollitraut et al, 2010  Comparison with Argo floats in the Atlantic over 2006/2007 period at 1000m  Comparison with current meters from RAPID- WATCH MOC array [Cunningham et al., 2007] in the western boundary current at 25°N in the Atlantic 26.5°North 76.5°West skill score [Taylor,2001]: increases when the standard deviation of the difference decreases and the correlation increases. The score penalizes methods with low variability. APR MAY JUN JUL AUG SEP OCT NOV DEC JAN FEB MAR APR 20042005 40m 100m From the CORIOLIS center T(z), S(z) From the NCDC center – Reynolds ¼° Sea Surface Temperature From the SSALTO/DUACS center Sea Level Anomaly Mean Dynamic Topography From the RIO05 [Rio and Schaeffer, 2005]SURCOUF3D GLORYS RAPID (current meters) RAPID (T/S + current meter at 100m through the thermal wind equation)SURCOUF3D_v2 40m 100m 400m 800m 1200m References Conclusion / Perspectives Application Zonal velocity (cm/s) Meridional velocity (cm/s)  Step 1 : extract the ‘steric’ part of SLA (Dhomps et al., 2010) + vertical projection of satellite SLA+SST data using a multiple linear regression method and covariances calculated from historical data  synthetic fields  Step 2 : combination of synthetic and in-situ T/S profiles using an optimal interpolation method  combined fields  Step1 and Step 2 fully described in Dhomps, 2010  ARMOR3D temperature/salinity fields : [0-1500m] – 1/3° grid - weekly  Combination with T/S profiles through the thermal wind equation ARIVO climatology T/S Climatology Rms difference (% variance) Temperature Rms difference (% variance) Salinity  Validation of step 1 over the year 2007 - using independent T/S profiles Years 2007 (~45 000 profiles)  Large improvements compared to previous estimate thanks to Arivo climatology + new covariances Years 2002-2008 (~3400 profiles) Rms difference (% variance) Global - Temperature Rms difference (% variance) Global - Salinity  Contribution of the Argo observing system visible at all depth, 10 to 20 % of the signal variance  Validation of step 2 over the years 2002- 2008 - using independent T/S profiles WOA05 Arivo Old New WOA05 Arivo Synth Combined Application  Temperature Variability patterns over the 2004-2008 period:  Synthetic fields, ARMOR3D, SCRIPPS (Roemmich et al., 2009) and SODA (Carton et al., 2008) show very similar patterns for temperature. Salinity patterns in SODA are more noisy (see additional material). ARMOR3DSCRIPPS SODA 2.2.4 2004 2008 Synthetic fields 2005 2006 2007  Moonitoring the Atlantic Meridional Overturning Circulation (AMOC) SURCOUF3D GLORYS RAPID APR MAY JUN JUL AUG SEP OCT NOV DEC JAN FEB MAR APR 20042005 Surcouf3D  Good consistency between Surcouf3D and current meters for zonal velocities at all depths and for the meridional velocities from the surface to 800m Surcouf3D  Surcouf3D does not resolve the current inversion at 1200m because Armor3D is too smooth. Using T/S from Rapid, that have a better resolution, permits to resolve the inversion. FloridaAfrica Surcouf3D Comparison of 3 different methods to compute the maximum AMOC strength in the Atlantic at 26.5°N. The first method uses in-situ data from the RAPID-WATHCH MOC array. Surcouf3D uses a combinaison through the thermal wind equation of in-situ and remote-sensing observations. GLORYS is a numerical model reanalysis.  Good consistency between the three different fields.  High variability  Intercomparison useful for the cross validation of the various datasets as well as for assessing the error on the MOC computation Transport (SV) Standard deviation (cm/s) Correlation coefficient 400m 800m 1200m