Satellite radar altimetry of sea ice

The thesis concerns the analysis and interpretation of data from satellite borne radar altimeters over ice covered ocean surfaces. The applications of radar altimetry are described in detail and consider monitoring global climate change, the role that sea ice plays in the climate system, operational applications and the extension of high precision surface elevation measurements into areas of sea ice. The general nature of sea ice cover is discussed and a list of requirements for sea ice monitoring is provided and the capability of different satellite sensors to satisfy needs is examined. The operation of satellite borne altimeter over non-ocean surfaces is discussed in detail. Theories of radar backscatter over sea ice are described and are used to predict the radar altimeter response to different types of sea ice cover. Methods employed for analysis of altimeter data over sea ice are also described. Data from the Seasat altimeter is examined on a regional and global scale and compared with sea ice climatology. Data from the Geosat altimeter is compared with co-incident imagery from the Advanced Very High Resolution Radiometer and also from airborne Synthetic Aperture Radar. Correlations are observed between the altimeter data and imagery for the ice edge position, zones within the ice cover, new ice and leads, vast floes and the fast ice boundary. An analysis of data collected by the Geosat altimeter over a period of more than two years is used to derive seasonal and inter-annual variations in the total Antarctic sea ice extent. In addition the retrieval of high accuracy elevation measurements over sea ice areas is carried out. These data are used to produce improved maps of sea surface topography over ice- covered ocean and provide evidence of the ability of the altimeter to determine sea ice freeboard directly. In addition the changing freeboard of two giant Antarctic tabular icebergs, as measured by the Geosat altimeter, is presented. As a summary the achievements are reviewed and suggestions are made towards directions for further work on present data sets and for future data from the ERS-1 satellite