Summary: | It is essential for winter shipping in the Baltic Sea to get reliable and up-to-date information of its rapidly changing ice conditions. Spaceborne synthetic aperture radar (SAR) images are the only way to produce this information operationally in fine scale independent of daylight and nearly independent of weather conditions. Currently, classification algorithms for the RADARSAT-1 and ENVISAT SAR images utilize mainly the image structure and only limited information on sea ice geophysics and empirical statistics of backscattering signatures of various ice types are utilized. Therefore, interpretation of the classification results is often difficult. Both classification results and their interpretation should very likely improve with the addition of this information. Spaceborne microwave radiometer data are not suitable for the operational Baltic Sea ice monitoring aiding ship navigation due to their coarse spatial resolution, but they can provide an independent data source on the sea ice conditions for validation of the SAR classification algorithms. Both SAR and radiometer data based sea ice products can also be utilized in the geophysical studies of the Baltic Sea ice. In order to support development of operational classification algorithms for SAR and radiometer data, basic research on the microwave remote sensing of the Baltic Sea ice has been conducted in this work. The research work included the following topics: (1) statistics of C- and X-band backscattering signatures of various ice types, (2) statistics of L- and C-band polarimetric discriminants of various ice types, (3) radar incidence angle dependence of backscattering coefficient (σ°) in RADARSAT-1 SAR images, (4) dependence between standard deviation and measurement length for σ° signatures and its usability in sea ice classification, (5) comparison between SAR σ° time series and results from a thermodynamic snow/ice model, and (6) statistics of passive microwave signatures of various ice types. Additionally, a comprehensive literature review of ...
|