 Researchers have used object-based image analysis to quantitatively link winter first-year sea ice surface roughness to C-band radar SAT-2 and L-band ALOS-2 CalSA-2 SAR backscatter measured in the Canadian Arctic. Results show that C-band HH polarization backscatter is correlated with roughness at a shallow incidence angle while L-band HH and VV polarization backscatter is correlated with roughness at a shallow incidence angle. Retrieval models for surface roughness are developed and applied to the imagery, demonstrating the potential of SAR for mapping deformation with a best-case RMSE of 5 mm, strong correlations between C- and L-band SAR backscatter acquired during the advanced melt period and winter measured surface roughness are observed, suggesting that SAR potentially provide significant sea ice structural information despite the occurrence of melt and presence of surface melt ponds. This article was authored by Sylvie Marie Caffarella, Randall Sherian, Torsten Gelsetzer, and others.