TRACKING OF ICE EDGES AND ICE FLOES BY WAVELET ANALYSIS OF SAR IMAGES

This paper demonstrates the use of wavelet transforms in the tracking of sequential ice features in the ERS-1 synthetic aperture radar (SAR) imagery, especially in situations where feature correlation technique s fail to yield reasonable results, Examples include the evolution of the St. Lawrence polynya and summer sea ice change in the Beaufort Sea . For the polynya, the evolution of the region of young ice growth sur rounding a polynya can be easily tracked by wavelet analysis due to th e large backscatter difference between the young and old ice. Also wit hin the polynya, a 2D fast Fourier transform (FFT) is used to identify the extent of the Langmuir circulation region, which is coincident wi th the wave-agitated frazil ice growth region, where the sea ice exper iences its fastest growth. Therefore, the combination of wavelet and F FT analysis of SAR images provides for the large-scale monitoring of d ifferent polynya features. For summer ice, previous work shows that th is is the most difficult period for ice trackers due to the lack of fe atures on the sea ice cover. The multiscale wavelet analysis shows tha t this method delineates the detailed flee shapes during this period, so that between consecutive images, the flee translation and rotation can be estimated.