A study of the scatterometer image reconstruction algorithm and its applications to polar ice studies.

摘要

A spaceborne microwave scatterometer is designed for the remote sensing of near-surface ocean winds, but rapid repeat coverage and immunity to meteorological effects makes it ideal for studying polar ice as well. Unfortunately, the intrinsic resolution of the scatterometer is insufficient for many geophysical studies.; Using multiple, overlapping passes of the scatterometer, the scatterometer data can be reconstructed into enhanced resolution images with resolutions better than the intrinsic 25-50km instrument resolution. This research establishes a theoretical backdrop for resolution enhancement of scatterometer data using the irregular sampling characteristics of multiple passes of the scatterometer. An optimization of the previously developed SIR algorithm for the ERS-1 scatterometer is presented. Issues of algorithm initialization, cubic estimation and acceleration of the iterative updates are addressed. The work is concluded with an application of the SIR resolution enhancement algorithm to Southern Ocean sea ice, including an examination of azimuthal modulation of backscatter and a simple sea ice classifier.