We propose to apply SAR (synthetic aperture radar) imagery in a novel way to measure the thickness of the frazil and pancake ice which is found to be the chief component of the odden ice tongue in the Greenland Sea. In this way we can assess whether the salt flux from local ice production is sufficient to trigger deep convection in winter.
The SAR application makes use of the discovery that ocean waves and swell propagate into frazil and pancake icefields in such a way that their wavefronts are visible on SAR images, but suffer a change of wavelength (and thus of direction) which is dependent on the thickness of the frazil-pancake cover. The change in wavelength and wave direction can then be used to infer ice thickness.
The project involves systematically analysing ERS-1 SAR imagery from the winter period in the Greenland Sea during which the Odden ice tongue exists (usually January-May), and for the years 1993 and 1994. Wind data are used to establish the dominant incident wave direction, then small (5 x5km) box samples are taken from each image along an incoming wave vector, and subjected to Fourier analysis to yield wavelength and direction The results are interpreted in the light of a theoretical model of wave propagation in an icefield of non-interacting mass points, and used to calculate ice thickness. The variation in space and time through the winter, in conjunction with satellite data on odden ice extent obtained as part of an EC MAST-II programme, are used to estimate salt fluxes due to ice production. These results are applied to current models of the convection process to test whether the salt fluxes observed are sufficient to trigger convection. Hydrographic data, available during both years, will test whether convection indeed took place. If time permits, the 1992 Odden will also be studied.
The project also involves further refinement of the wave dispersion theory upon which the technique is based.
Funding SchemeCSC - Cost-sharing contracts