On the Sea Ice Motion Estimation with Synthetic Aperture Radar

Kurzfassung

Remote sensing provides reliable information on sea ice conditions and trends. Space born synthetic aperture radar (SAR) is a key sensor due to its high spatial and radiometrical resolution, large coverage area and all weather capability among other advantages. From a scientific perspective the sea ice drift derived from SAR data helps understanding the sea ice kinematics and dynamics on a fine scale. On the other hand the information on drift is used to support maritime safety and offshore engineering for oil and gas industry in the Polar Regions. A high resolution sea motion estimation algorithm from a sequence of SAR images is presented. Similar to previous approaches developed for C-band phase correlation within a hierarchical motion estimation framework. Unlike previous algorithms, the framework is able to deal with discontinuities in the motion field. The algorithm is applied to a time series of TS-X ScanSAR Mode as well as TS-X ScanSAR -ENVISAT Image Mode data pairs acquired from different orbits over the Arctic Sea. Validation is done by computing the cross correlation coefficient after image warping. Results indicate good performance in terms of tracking of pack ice. Moreover the algorithm is able to correctly estimate the drift in regions of discontinuous motion. Furthermore the algorithm is capable of combining X-band and C-Band SAR imagery and is therefore useful for filling coverage gaps as well as observing fast ice dynamics on shorter time scales when different SAR frequency bands and missions are utilized.