First-year sea ice melt pond fraction estimation from dual-polarisation C-band SAR - Part 2: Scaling in situ to Radarsat-2

Sea ice melt pond fraction (f(p)), linked with lower sea ice surface albedo and increased light transmittance to the ocean, is inadequately parameterised in sea ice models due to a lack of observations. In this paper, results from a multiscale remote-sensing program dedicated to the retrieval of level first-year sea ice (FYI) f(p) from dual co-and cross-polarisation C-band synthetic aperture radar (SAR) backscatter are detailed. Models which utilise the dominant effect of free-water melt ponds on the VV/HH (vertical transmit and vertical receive/horizontal transmit and horizontal receive) polarisation ratio at high incidence angles are tested for their ability to provide estimates of the subscale fp. Retrieved f(p) from noise-corrected Radarsat-2 quad-polarisation scenes are in good agreement with observations from coincident aerial survey data, with root mean square errors (RMSEs) of 0.05-0.07 obtained during intermediate and late stages of ponding. Weak model performance is attributed to the presence of wet snow and slush during initial ponding, and a synoptically driven freezing event causing ice lids to form on ponds. The HV/HH (horizontal transmit and vertical receive/horizontal transmit and horizontal receive) ratio explains a greater portion of variability in fp, compared to VV/HH, when ice lids are present. Generally low HV channel intensity suggests limited applications using dual cross-polarisation data, except with systems that have exceptionally low noise floors. Results demonstrate the overall potential of dual-polarisation SAR for standalone or complementary observations of f(p) for process-scale studies and improvements to model parameterisations.