Land Surface Properties near Terra Nova Bay, East Antarctica, Analyzed by Time-series Height, Coherence and Amplitude Maps Derived from COSMO-SkyMed One-day Tandem Pairs

Abstract:

Analysis of the surface properties of Antarctica is very important to study the change of environment and climate in the polar region. Synthetic aperture radar (SAR) has been widely used to study Antarctic surface properties because it is independent of sun altitude and atmospheric conditions. Interferometric SAR (InSAR) observes surface topography and deformation, by calculating the phase differences between two or more SAR images obtained over same area. InSAR technique can be used for height mapping in stable areas with a few meter accuracy. However, the InSAR-derived height map can have errors if the phase differences due to surface deformation or change of the scattering center by microwave penetration into snow are misinterpreted as the elevation. In this study, we generated the height maps around Terra Nova Bay in East Antarctica from 13 COSMO-SkyMed one-day tandem InSAR pairs obtained from December 2010 to January 2012. By analyzing the height maps averaged over the 13 interferograms and its standard deviation (STD) map, we could classify the surface types into glacier, mountains and basin areas covered with snow. The mountain areas showed very small STD because its surface property is unchanged with time, except for the small STD values caused by the errors from the unwrapping processing, satellite orbit or atmospheric phase distortion. Over the basin areas, however, the STD of the height was much larger than the mountain area due to the variation of scattering center either from the change in surface property such as snowfall and sublimation or by the surface displacement of snow mass that are too slow. A year-long constant motion of such slow-creeping snow body was positively identified by its linear relationship between the misinterpreted elevation and the baseline perpendicular component of InSAR pair. Analysis of time-series coherence maps and amplitude maps have also contributed to clarify the surface properties and its changes due to various environmental factors such as snow fall, wind, sublimation, and the freezing-thawing processes in this Antarctic land surface.

Acknowledgement – This research was supported by National Research Foundation of Korea through NRF-2013R1A1A2008062 and NRF-2013M1A3A3A02041853.