Reduced ascending/descending pass bias in SMOS salinity data demonstrated by observing westward-propagating features in the South Indian Ocean

The European Space Agency (ESA) Soil Moisture and Ocean Salinity (SMOS) satellite has been providing data, including sea surface salinity (SSS) measurements, for more than five years. However, the operational ESA Level 2 SSS data are known to have significant spatially and temporally varying biases between measurements from ascending passes (SSSA) and measurements from descending passes (SSSD). This paper demonstrates how these biases are reduced through the use of SSS anomalies. Climatology products are constructed using SMOS Level 2 data to provide daily, one-degree by one-degree climatologies separately for ascending and descending passes using a moving window approach (in time and space). The daily, one-degree products can then be averaged to provide values of climatological SSS at different spatial and/or temporal resolutions. The averaged values of the SMOS climatology products are in good general agreement with data from the World Ocean Atlas 2013. However, there are significant differences at high latitudes, as well as in coastal and dynamic regions, as found by previous studies. Both the mean and standard deviation of the differences between data from ascending passes and data from descending passes for the anomalies are reduced compared with those obtained using the original salinity values. Geophysical signals are clearly visible in the anomaly products and an example is shown in the Southern Indian Ocean of westward-propagating signals that we conclude represent the surface expression of Rossby waves or large-scale non-linear eddies. The signals seen in salinity data agree (in speed) with those from sea surface temperature and sea surface height and are consistent with previous studies.