Meetings: Documents

ENSO Signature in the SMOS Sea Surface Salinity Maps
[03-Dec-2012] Ballabrera, J., Umbert, M., Hoareau, N., Turiel, A., and Font, J.
Presented at the 2012 AGU Fall Meeting
Until recently, the role of salinity observations in the operational simulation and prediction of ENSO was neglected because of the historical lack of observations and because leading intermediate coupled models had significant predictive skill without directly accounting for salinity effects. In Ballabrera-Poy et al., (2002), the potential role of sea surface salinity (SSS) observations on the statistical predictions of ENSO was investigated. It was shown that, although SSS observations would play little role in statistical nowcasts of ENSO, they would provide a significant role in the 6-12 month predictions.
The European Space Agency (ESA) Soil Moisture and Ocean Salinity (SMOS) Earth Explorer opportunity mission was launched on November 2, 2009, becoming the first satellite mission addressing the challenge of measuring SSS from space with the help of MIRAS (Microwave Imaging Radiometer with Aperture Synthesis), a novel two-dimensional interferometer operating at L-band (1.4 GHz). Although the L-band frequency is the optimal for ocean salinity measurements, the retrieval of SSS information requires special care because of the low sensitivity of the brightness temperature to SSS: from 0.2-0.8 K per salinity unit.

Maps of 10-day averages of SSS in 1x1 degree boxes are distributed by the SMOS Barcelona Expert Centre on Radiometric Calibration and Ocean Salinity (SMOS-BEC, These maps are derived from the SMOS reprocessing campaign released to the SMOS user community in March 2011, and span the period from January 2010 through December 2011. The current accuracy of these SSS maps ranges from 0.2-0.4, depending on the ocean region being considered (Umbert et al., 2012).
During the period of the reprocessing campaign, the equatorial Pacific has been in a quasi-continuous La Niña state. During the cold phases of ENSO, positive anomalies of SSS are expected with a largest anomalous values in the western warm-fresh pool. The anomalies derived from the SMOS data do indeed display a positive anomaly. The persistence of the feature, its geographical pattern, and the time modulation of the anomaly amplitude indicate that the SSS fields provided by the SMOS mission do actually contain an ENSO-related signal, despite the shortness of the data gathered by the mission. Although the SMOS time span is still too short to allow any study about the factual impact of these data in the prediction of ENSO (either through data assimilation or statistical predictions), these results provide a clear indication that the novel observational technology that SMOS represents for the observation of the Earth is currently able to capture seasonal and interannual signatures of climate interest.