On the Decorrelation Effect of Dry Snow in Differential SAR Interferometry

Plenty of data records demonstrate that differential InSAR acquisitions of snow covered areas are often affected by severe temporal decorrelation, complicating the estimation of snow physical parameters such as the snow water equivalent. The decorrelation effect is commonly attributed to a change in the underlying scattering center distribution due to melting/refreezing, compacting of snow, or redistribution of underlying vegetation. We demonstrate that a mere change of the dielectric constant of a dry snow cover may lead to severe decorrelation, even without a change in scatterer distribution, which provides additional opportunities for the estimation of snow parameters. In this paper, a first discussion of the snow-induced decorrelation effect is provided and the derived model is evaluated against Sentinel-1 12-day coherence data using SWE measurements provided by the Copernicus Global Land Service.