Enhanced Global Sea-Ice CFOSAT sigma₀ maps Reprocessing Utilizing HEALPix-Based Radar within the SROLL Framework

In this presentation we investigate sea ice physical parameters by undertaking an extensive reanalysis of radar remote sensing data from SWIM and SCAT sensors onboard the french-chinse CFOSAT satellite. The central objective is to estimate daily maps of sea ice extent, type and displacement from radar sigma_0 data which is linked with surface roughness at a spatial resolution of 12.5 km. For this purpose, it is needed to know biais of the sigma₀ maps, this is what will be presented here. A significant challenge in processing sea-ice data is handling observations concentrated near the poles, where noteworthy features exist, while systematic instrument effects are more stable and manageable at lower latitudes, such as over continents. To prevent biases from arising due to geographic projections, we apply the HEALPix pixelization, functioning as a Discrete Global Grid System (DGGS). This technique enables us to process the complete dataset at once, extracting both instrumental biases and the relevant signal within a cohesive framework. The map production employs SROLL, a methodology originally crafted for processing cosmology data in the Planck mission. SROLL is tailored for calibrating, denoising, and producing consistent maps in a single operation, utilizing all available satellite data. We processed five years of SWIM observations and two years of SCAT data in one run gathering as much as possible all available information. Temporal gaps, related to the scanning strategy, were filled using spline-based interpolation, and detected antenna gain variations were adjusted. Additionally, analyses and compensation were performed for long-term noise fluctuations. The resulting datasets underwent successful validation against independent references, illustrating the approach's robustness. This work underscores SROLL's paradigm efficacy in satellite data processing and emphasizes its potential across space missions beyond cosmology. The data is publicly available in Zarr format, promoting ease of access and compatibility with the xDGGS framework.