MONRUK SAR data analysis report

• Main Authors: Sandven, Stein, Hamre, Torill, Babiker, Mohamed, Kloster, Kjell, Hansen, Morten W., Wåhlin, Johan, Kudriavtsev, Vladimir, Myasoedov, A., Alexandrov, V. Y., Melentyev, V. V., Zakhvatkina, N.Y., Yarigina, A.O., Feoktistov, A.A., Novikova, N.N., Denisov, P.V., Malinovsky, V. V., Spivak, L.
• Format: Report
• Language: English
• Published: Zenodo 2009
• Subjects: Barents Sea; Kara Sea; Caspian Sea; ENVISAT; Synthetic Aperture Radar; ASAR; Ocean; Mesoscale; Front; Polynya; Iceberg; Iceberg*; Sea ice
• Online Access: https://doi.org/10.5281/zenodo.7520467

Part 1: SAR data analysis for open ocean. ENVISAT ASAR images for the three study areas, Barents and Kara Seas, Black Sea and Caspian Sea, are collected to investigate the most important oceanographic phenomena with use of existing and newly developed retrieval algorithms. ASAR data were received from the ESA rolling archive, stored at NERSC and NIERSC servers, and then distributed among the project partners. In most of the cases the set of ASAR images is analysed in combination with accompanying metocean data, supplemented with IR and optical satellite data, and data from oceanic numerical models. The main tools implemented for the SAR analysis is SARTool provided by BOOST Technologies and the Radar Imaging Model (RIM) developed at NIERSC. Part 2: SAR data analysis for sea ice. A series of ENVISAT ASAR WS images, covering the western part of the Barents and Kara seas, including the Barents and Kara Seas in the period from autumn of 2007 until May 2009 have been analyzed and visually interpreted. The following procedures of ENVISAT ASAR image preliminary processing were elaborated in order to improve and simplify their visual interpretation: recalculation of the backscatter coefficient to pre-defined incidence angle and automated processing of SAR mosaics. A neural network algorithm of sea ice classification was developed and tested for the images over the Central Arctic. The algorithm of multiyear ice concentration retrieval consisted of sea ice classification using Bayesian approach, interactive delineation of zones of different partial concentration of multiyear ice and calculation of multiyear ice concentration within the boundaries of delineated areas. An algorithm of ice drift, based on cross-correlation technique, was developed and tested using a series of ENVISAT ASAR images. NERSC Technical Report no. 310. Funded by EU, FP6 Contract no. 031001