Polarimetric analysis of snow-covered and bare lake ice from Ku and X-band scatterometer data

Lake ice plays a key role in regional climate, and has significant physical, biological and socio-economic impacts (e.g. fish overwintering habitat, winter-road transportation, public safety). In the last two decades, there has been growing interest by the international remote sensing community to explore radar polarimetry for glaciological investigations, mainly for glaciers and ice sheet. Polarimetric synthetic aperture radar (SAR) could be a potential tool for lake ice cover mapping and ice thickness estimation. In this paper, we represent results from the first investigation of fully polarimetric Ku and X-band (9.6 and 17.2 GHz, respectively) scatterometer data collected over lake near Churchill, Manitoba. Several controlled and calibrated experimental measurements were carried out during winter 2010-2011, as a contribution to the Cold Regions Hydrology High-resolution Observatory (CoReH2O) candidate mission of the European Space Agency (ESA). Scatterometer scans were made on several occasions at five undisturbed static sites on Ramsey Lake. Measurements characterizing snow and ice properties were also gathered immediately after scatterometer scans. Snow depth and density, snow water equivalent, gain size, ice thickness, ice composition and air inclusion in ice volume were determined at each site. This field data set was very important for the interpretation of the polarimetric parameters, e.g. the copolarization ratio, the copolarization phase and the depolarization ratio. First, the polarimetric parameters have been analysed for the two layers (snow and ice) covariance matrix and where snow subsequently removed. Thus, the influence of the snow layer on the polarimetric data could be quantified. Also, the Pauli and Cloude/Pottier polarimetric decompositions were applied for the two-layer and one-layer scattering mechanisms (removed snow) to quantify the effectiveness of these decompositions. Results show that the polarimetric SAR could explain the different complex scattering mechanism of the snow-covered lake.