Sea ice radar backscatter modeling, measurements, and the fusion of active and passive microwave data

The polar oceans play a key role in our climate because of the complex interactions between sea ice, open water and the atmosphere. Because of the persistent cloud cover and darkness in the Arctic for roughly half the year, the most viable means for monitoring sea ice on a global basis is with satellite microwave remote sensing instruments. The development of techniques for utilizing microwave remote sensing data depend on our understanding of the backscatter and emission properties of sea ice. Seasonal transitions are particularly difficult for interpretation of microwave scattering and emission from sea ice, due to the rapidly changing environmental conditions.; We investigated several key issues concerning the radar backscatter signatures of Arctic sea ice during the summer-to-fall transition and examined the potential for fusing active and passive microwave data for ice type concentration estimation.; To improve measurements made with short-range FM and step-frequency radar systems we developed coherent noise reduction techniques. We have demonstrated that these techniques are useful for a number of the systems designed for exploring the backscatter properties of sea ice. These coherent noise reduction methods improve the sensitivity of single-antenna radar systems by as much as 25 dB.; We obtained C-band measurements of radar backscatter from multiyear ice during the onset of freeze-up in the Arctic. Through the use of measurements of ice and snow physical properties and electromagnetic scattering models we determined the scattering mechanisms responsible for the observed changes in radar backscatter. The refreezing of moisture on or near the ice surface of multiyear ice causes a decrease in the lossy part of the permittivity of the ice. This allows more energy to penetrate into the sea ice volume, which results in increased volume scattering. This results in a dramatic increase in the radar backscatter from sea ice at the onset of freeze-up. We demonstrated that these changes are also observed on a large scale from satellite synthetic aperture radar (SAR) data and can be used to detect the onset of freeze-up in the Arctic.; Finally, we utilized a technique to combine active and passive microwave data for estimating ice type concentration during the freeze-up season. This technique uses multiyear ice concentrations, obtained by analysis of ERS-1 SAR data. The multiyear ice concentration, derived from the SAR data, is used to constrain a multispectral algorithm for determining ice type concentration from satellite passive microwave data. According to ship-based ice observations the "fused" estimates of first-year ice concentration appear to be more accurate than estimates based on the SSM/I data alone.