Microwave remote sensing of young sea ice in the Greenland Sea

During winter 1993 data from satellite remote sensing and two research cruises showed the Odden ice tongue in the Greenland Sea to consist predominantly of young sea ice in the form of frazil and pancake ice. These young ice types have been implicated in the preconditioning of the ocean for deep ventilation. Satellite-borne microwave instruments potentially offer the ability to monitor important ice-cover parameters. A dataset consisting of Synthetic Aperture Radar (SAR) and Special Sensor Microwave Imager (SSM/I) imagery is used in conjunction with ship-based measurements to improve the accuracy of satellite classification techniques. The high resolution SAR data is compared with an under-way video log taken from the bridge of a ship during two days of fieldwork in April 1993. Bands of frazil and changing concentrations of pancake ice are identified and calibrated backscatter values at a range of incidence angles obtained. It is shown that separating pancake ice and open water is difficult. SAR data is then used to identify areas of frazil/new ice formation in the SSM/I dataset. The NASA-Team algorithm, commonly used to estimate ice concentrations, is shown to classify this new ice as open water. The new ice is also seen to alter the SSM/I microwave signature in a fashion characterised by a high polarisation difference. A principal component analysis and the depolarising effect of wind on the open-water signature are used to argue for a different ocean mask. This ocean mask is shown to leave areas of new ice which may be distinguished using a further new-ice mask. Ocean waves propagating into the ice cover are investigated using SAR data, with the intention of linking changes in wavelength to ice thickness. Results are presented from using ice spectra in a nonlinear inversion algorithm, previously used for ocean spectra. A quasi-linear model is then used to indicate that the ice cover extends the azimuthal cut-off to higher wavenumbers and this alone can alter the detected peaks in the SAR spectra. Finally, a linear analysis returns unlikely values for the elastic properties of the ice cover.

Name dc:type.qualificationname

Doctor of Philosophy (PhD)

Level dc:type.qualificationlevel

Doctoral

Grantor dc:publisher.institution

University of Cambridge

Year dc:date.issued

1997

Author dc:creator

• Tadross, Mark Alexander

Licence

• https://apollo8-f-pro.lib.cam.ac.uk/bitstreams/aef92cef-96f7-409a-946e-e21f04398ee6/download
• http://purl.org/NET/rdflicense/allrightsreserved/

DOI dc:identifier.doi

https://doi.org/10.17863/CAM.116103

OAI identifier oai:identifier

oai:www.repository.cam.ac.uk:1810/380527