Supervisors:
DSc Mart Noorma, University of Tartu, Estonia
PhD Rein Rõõm, University of Tartu, Estonia
Opponents:
PhD Liis Sipelgas, Tallinn University of Technology, Estonia PhD Laurent Ferro-Famil, University of Rennes, France
Summary:
Synthetic Aperture Radar (SAR) is a land and water surface remote sensing instrument typically used on aeroplanes and satellites. SARs work in radio and microwave spectral regions with wavelengths from 1 m to 3 cm and are sensitive to sensed objects structure and electrical properties. In the current thesis X-band SAR applications for flood mapping in forest and grassland parameters retrieval are tested. The tests done during spring 2010 have proven X-band SAR suitability for flood detection in Northern European temperate forest during leaf-off season. Before this work it was commonly believed that X-band SAR forest penetration is not enough to detect water under forest canopy. The improvement of using HH-VV polarimetric channel over conventional HH for flood detection in forest was measured. HH-VV channel provided 0.2 to 1.6 dB higher flooded vs non-flooded forest backscatter based distinction than conventional HH channel. In grasslands X-band SAR was able to reveal the areas with freshly cut uncollected grass according to the tests carried out in Matsalu grasslands in summer 2011. The regions with freshly cut uncollected grass corresponded to dominant alpha parameter of about 25 degrees, whereas for other grassland states the same parameter was around 10 degrees.