Toward a CryoSat-2 / Sentinel-3 continuum of sea-ice thickness and volume observations

Antoine Laforge (LEGOS, France)


Sara Fleury (LEGOS/CTOH, France); Kévin Guerreiro (LEGOS, France); Florence Birol (LEGOS, France); Salvatore Dinardo (EUMETSAT, Germany); Giovanni Sabatino (ESRIN, Italy); Jérôme Benveniste (ESRIN/ESA, Italy); Jérôme Bouffard (ESRIN/ESA, Italy)

Event: 2018 Ocean Surface Topography Science Team Meeting

Session: Instrument Processing: Measurement and Retracking

Presentation type: Type Oral

Arctic sea-ice plays a major role on the global climate and is a key witness to the on-going climate change, which justifies the importance of closely monitoring its evolution. While sea-ice extent is available since the premises of satellite observations, its thickness and volume are still poorly known.
The quite recent SAR delay-Doppler in altimetry has brought new possibilities in ice thickness estimation. The pioneer of this technic is SIRAL on-board CryoSat-2, which continuously observed the Polar Regions since 2010, providing the first sea-ice thickness time series with good accordance with in-situ measurements. The recent launch of Sentinel-3A on 16 February 2016, and the planned launch of Sentinel-3B before the end of 2018, both equipped with a SAR altimeter, promises for a long record of the sea-ice up to 81.5 degrees of latitude. Nevertheless, the continuity of the sea ice thickness with CryoSat-2 still raised some difficulties and no solution has been published yet. The Sentinel-3 and CryoSat-2 altimeters being very similar the problem most probably comes from the ground processing, which differ because of their different objectives: the CryoSat-2 Earth Explorer scientific mission is dedicated to the observation of the Polar Regions whereas the Sentinel-3 Copernicus operational missions focused on lower latitudes.
The ground processing differs at distinct, the main ones being the Doppler SAR processing, that provides the synthetic echoes, and the retracking, that extracts the range from the echoes. In the presented study, we have considered these two levels and analysed step by step the impacts of the different choices. For the Doppler SAR processing, two options present in the CryoSat-2 ground processing chain have been deactivated in the Sentinel-3: the hamming filtering and the zero-padding interpolation. Thanks to the configurable SARvatore processing chain running on the ESA - Research and Service Support (RSS) processing platform, we have been able to analyse the impact of both of these options on the sea-ice thickness measurements and to conclude on the configuration that provides the best results. In a similar process, we have compared the effects of two retrackers belonging to two different families: TFMRA, a robust but heuristic threshold retracker, and SAMOSA+, a retracker based on a physical model that can handle the very atypical specular waveforms over oceans that are encountered over sea-ice.
For a better understanding of the effects of these various options, we have evaluated separately their impacts on the diffuse waveforms, encountered over the ice floes, and on the specular waveforms that are specific to the ones encountered over the leads (water in the sea-ice fractures).
The resulting sea-ice freeboards have been evaluated against different datasets of in-situ measurements (OIB, IMB, BGEP).
Finally, these configurations have been applied on the full December 2016 month of data over the Arctic Basin up to 81.5°N with CryoSat-2 on one hand, and Sentinel-3 on the other hand. The mean difference between the output freeboard maps is of 0.3 cm with a standard deviation of 5.4cm. The relatively high standard deviation may come from the large proportion of thin ice still present in December, and which is difficult to measure by altimetry. Other months of winter will be analysed for the final presentation, but these results already demonstrate the ability of Sentinel-3 to ensure the continuity with CryoSat-2 of the sea-ice thickness and volume observations.
Acknowledgments: This work is funded by the ESA CryoSeaNICE project. It could not have been accomplished without the support of ESA - RSS and SARvatore SAR processing chain.

Oral presentation show times:

Room Start Date End Date
Teatro Auditorium Thu, Sep 27 2018,16:15 Thu, Sep 27 2018,16:30
Antoine Laforge