HR: 1330h
AN: C43A-04    [Abstracts]
TI: Classification of first-year sea ice deformation using ENVISAT ASAR AP data
AU: * Breneman, C
EM: cbrenema@ucalgary.ca
AF: University of Calgary, Department of Geography 2500 University Drive N.W. , Calgary, AB T2N 1N4 Canada
AU: Yackel, J J
EM: yackel@ucalgary.ca
AF: University of Calgary, Department of Geography 2500 University Drive N.W. , Calgary, AB T2N 1N4 Canada
AB: Sea ice deformation is an important factor in sea ice thickness distribution and mass balance studies. Deformation identification and characterization is also crucial for safe winter ship navigation in Arctic regions. This research evaluates the utility of individual and multiple ENVISAT ASAR system parameters and texture parameters to discriminate between smooth, rough and deformed first year Arctic sea ice. Smooth, rough and deformed classes correspond to unique insitu measured sea ice topography values and SAR backscatter properties. System parameters; single polarizations (HH, VV, HV and VH), dual polarization combinations (HH + VV, HH + HV and VV + VH) and incidence angle (15.0 to 45.2 degrees) and texture measures appropriate for SAR imagery (contrast, entropy and correlation) are investigated. Univariate and multivariate analyses are conducted using a minimum distance classifier to determine which individual or set of multiple parameters maximize the discrimination of the three sea ice topography classes. Our results indicate that increasing incidence angle results in an increased ability to discriminate sea ice deformation classes regardless of polarization combination or texture measure. For single polarizations, co-polarizations (HH & VV) outperformed cross-polarizations (HV & HH) by 6.48 % to 33.93 %. For dual polarizations, HH + VV polarizations outperformed VV + VH and HH + HV by 3.75 % to 14.93 %.
DE: 1594 Instruments and techniques
DE: 1635 Oceans (4203)
DE: 1640 Remote sensing
DE: 1863 Snow and ice (1827)
SC: Cryosphere [C]
MN: 2005 Joint Assembly