SAR image wave spectra to retrieve the thickness of grease-pancake sea ice using viscous wave propagation models

Young sea ice composed of grease and pancake ice (GPI), as well as thin floes, considered to be the most common form of sea ice fringing Antarctica, is now becoming the "new normal" also in the Arctic. A study of the rheological properties of GPI is carried out by comparing the predictions of two viscous wave propagation models: the Keller model and the close-packing (CP) model, with the observed wave attenuation obtained by SAR image techniques. In order to fit observations, it is shown that describing GPI as a viscous medium requires the adoption of an ice viscosity which increases with the ice thickness. The consequences regarding the possibility of ice thickness retrieval from remote sensing data of wave attenuation are discussed. We provide examples of GPI thickness retrievals from a Sentinel-1 C band SAR image taken in the Beaufort Sea on 1 November 2015, and three CosmoSkyMed X band SAR images taken in the Weddell Sea on March 2019. The estimated GPI thicknesses are consistent with concurrent SMOS measurements and available local samplings.