ABSTRACT Results obtained with a unique dual-antenna impulse radar system used to profile first- and multi-year sea ice near Prudhoe Bay, Alaska, are discussed. A description of the radar system is given along with representative field data. Continuous ice thickness profiles are required for studies related to modeling and understanding the dynamics of the sea ice cover, heat exchange between the ocean and the atmosphere and mass balance of the ice cover. Ice thickness profiles are also required for studies related to the determination of ice load distributions and force analyses, and the subsurface roughness as it pertains to under-ice sound propagation and to the quantity of oil that would be trapped in the undulating bottom relief of first-year and multiyear ice should an oil blowout occur in an ice covered sea. From the radar impulse travel times obtained with the use of dual antennas, calculations of thickness, electromagnetic impulse velocity and effective dielectric constant of the ice were made. Ice thicknesses determined by direct measurement and those calculated using the radar impulse travel times were found to be in good agreement. Continuous ice thickness profiles obtained with the radar were analyzed to provide representative cross sections of first-year and multi-year sea ice. These cross sections reveal the undulating bottom surface relief of both ice types. Calculations are presented that indicate a significant amount of oil could be trapped within this bottom relief should the oil be released under the ice from a sea-floor oi1production system. INTRODUCTION ' There is need for a system that can measure the top and bottom relief of both first-year and multi-year sea ice continuously, not only from the ice surface, but preferably at high speed from the air. This information is required for studies related to modeling and understanding the dynamics of the sea ice cover, heat exchange the ocean and the atmosphere, mass balance of the ice cover, ice load distributions and force analyses, and subsurface roughness as it pertains to under-ice sound propagation, marine life habitats and oil entrapment. Numerous ways of determining sea ice types and thicknesses have been tried in recent years, for example, visual observation, aerial photography, side-looking airborne radar (SLAR imagery), laser profi10meters, drop electrometers and sonar. Visual observations and aerial photography, such as those obtained on the U.S. Navy Birdseye sea ice reconnaissance flights, are highly valuable for compiling data on ice type and number of ridges and open leads per unit length. SLAR imagery results are highly variable and interpretation is difficult; for example, radar returns can indicate thin ice when in fact thick multi-year ice exists. Laser profile meter data are useful for determining such ridge statistics as number of ridges per unit length of track, surface roughness versus location, and ridge height distribution.