Research On Sea Ice Thickness Detection By Polarimetric Sar In Bohai Sea

Bohai Sea and its coastal regions are significant economic development areas in China. When winter comes, sea-water-frozen ice and its drifting movement pose a great threat to coastal construction and manufacturing industry, such as marine transportation, off-shore oil and gas exploration and fishery industry, leading to severe economic loss to our country. The evaluation of ice thickness and ice condition plays a crucial role in sea ice detection and surveillance. So far, the detection of sea ice thickness all around the world mainly depends upon in-situ observation. However, ice thickness retrieval that utilizes remote sensing technology, is still at experimental stage, and is immaturely operational.Polarimetric Synthetic Aperture Radar (PolSAR) not only has the ability of all weather and day and night measurement at which common SAR is competent, but can provide a wealth of polarimetric information, enabling it to be an ideal sea ice detection data source. In this dissertation, PolSAR data are used to study the sea ice thickness of Bohai Sea. Two methods are also developed, which are sea ice thickness level detection by ice classification method and sea ice thickness quantitative retrieval approach. In addition, a novel technology to detect sea ice thickness of Bohai Sea is initially put forward.As for the extraction of sea ice thickness level by sea ice classification: firstly, the ice types of Bohai Sea that can be identified by C-band SAR imagery is ascertained, associating the sea ice types with the sea ice thickness level of Bohai Sea, and establishing thickness-related ice classification system of space-borne C-band SAR; secondly, microwave-responding characteristics of sea ice are analyzed by SAR imagery, acquiring the classification system on the basis of this thesis. Sea ice types of Bohai Sea can be identified through SAR imagery, distinguishing ice thickness level of Bohai Sea. The validity of the classification system is substantiated. Thirdly, according to previous section, the polarimetric scattering characteristics of sea ice are extracted by high-resolution and fully-polarimetric SAR imagery. By combining the binary-tree classification idea, a sea ice classification algorithm of high-resolution and fully-polarimetric SAR data is developed, enhancing the sea ice classification accuracy by SAR. Finally, sea ice thickness level detection by SAR is realized on the basis of ice categorization result and our thickness-related ice classification system.As to the quantitative retrieval of ice thickness: in the first place, an experiment on microwave scattering characteristics of sea ice is conducted. Also, an electromagnetic scattering model of a different-distribution and multi-discrete scatter (air bubbles and brine inclusions) is exploringly studied, providing a theoretical instruction for a better understanding of sea ice electromagnetic scattering characteristics. In the second place, sea ice electromagnetic scattering model is combined with thermo-dynamic ice growth model of Bohai Sea to study and analyze the influences of different bands, polarization and incidence angles on the SAR capability of ice thickness retrieval, enriching the reference foundations for the following quantitative retrieval of ice thickness. Last but not least, by virtue of the electromagnetic theory of sea ice, the volumetric scattering model of Freeman decomposition is improved, putting forward a polarization-decomposition-based approach for PolSAR quantitative ice thickness retrieval. Experimental data are taken advantage to examine the method proposed in this dissertation.