Sea-ice thickness is a key quantity for many applications including climate research, weather forecasting, shipping and offshore. The capability to measure the thickness from satellite is of great importance given the vast area of the Arctic and Antarctic sea-ice cover. ESA’s CryoSat altimeter mission allows to measure the sea ice elevation (freeboard) above the sea level and in turn to estimate its thickness. This method works well for thick ice but bears large relative uncertainties for thin sea-ice because of the small freeboard. The spatial coverage of the altimeter measurements is confined along a narrow sub-satellite track and considerable averaging is required to reduce the inherent radar noise. For this reason CryoSat provides only limited information about the highly dynamic marginal ice zones (MIZ). On the other hand, ESA’s SMOS radiometer provides a complementary sensitivity for thin ice and complete daily coverage of the MIZ thus enabling a beneficial synergy of both missions. One main objective of the STSE SMOS+ Sea Ice study was therefore to develop a data fusion product based on CryoSat and SMOS data, now termed the CS2SMOS product. As a prerequisite, SMOS sea ice thickness retrieval methods had to be improved and validated with new field campaign data. Finally, an evaluation of the impact of the new SMOS observations on ocean-ice forecasts was performed.

A brief summary of main findings and results is given in the following:

• The improved SMOS sea ice thickness product, specified as version 3, is based on homogeneously reprocessed SMOS L1C brightness temperatures (TB) version v620. The new flags introduced with the version v620 of SMOS L1C TBs detect large parts of RFI contaminations and reduce data loss compared to previous versions. An evaluation of external noise sources (Faraday rotation, ascending/descending node inhomogeneities, galactic noise, sun glint) indicates no need for further TB corrections.
• Potential areas for improvement of the SMOS ice thickness retrieval algorithm and associated parameterizations have been investigated. We could not confirm significant advantages either for a method using a multi-layer radiation model with vertical ice temperature profile or for a method using an effective ice temperature instead of bulk ice temperature. A considerable uncertainty remains with respect to vertical salinity profile parameterizations. An evaluation of the effect of ice concentration on the thickness retrieval revealed only a weak dependency in the MIZ. The statistical correction for the thickness distribution was improved using a polynomial fit instead of a look-up table thereby eliminating discretization artefacts in the resulting thickness histogram.
• A dedicated validation campaign was conducted in the Barents Sea in March 2014. Thickness measurements from the ice strengthened research vessel Lance, a helicopter based on Lance, and the research aircraft Polar 5 operated from Svalbard airport formed an extensive and unique validation data set including measurements with the EMIRAD-2 L-band radiometer. The validation of two different SMOS sea ice thickness products confirms that the overall main pattern of the spatial thickness distribution is well captured. The thickness of deformed ice was considerably underestimated but the extensive areas of newly-grown young sea-ice were found in good agreement with the shipborne measurements
• 4) A merged synergy product of complementary weekly Arctic sea-ice thickness data records from the CryoSat altimeter and SMOS radiometer was developed. Based on an optimal interpolation (OI) scheme, a weekly Arctic-wide sea-ice thickness data set was generated. The benefit of the data merging is shown by a comparison with airborne thickness data in the Barents Sea. The synergy reveals a reduced root mean square deviation of about 0.7 m compared to the CryoSat retrieval and therefore demonstrates the great improvement in thin ice regimes.
• The impact of assimilating SMOS thin ice thickness data into the coupled ocean-sea ice data assimilation system TOPAZ was evaluated. TOPAZ is the operational Copernicus Arctic forecast system, which assimilates sea surface temperature (SST), altimetry data, temperature and salinity profiles, ice concentration, and ice drift with the ensemble Kalman filter (EnKF). Two parallel Observing System Experiments have been performed in March and November 2014, in which the SMOS sea ice thickness (thinner than 0.4 m) are assimilated in addition to the standard observational data sets. Validation against independent observations of ice thickness from buoys and ice draft from moorings indicates that there are no degradations in the pack ice but some improvements near the ice edge. The areas of largest impact are the Kara Sea, Canadian Archipelago, Baffin Bay, Beaufort Sea and Greenland Sea. The study suggests that the SMOS sea ice thickness is a good complementary data set that can be safely included in the Copernicus Marine Environmental Monitoring Services.