MEASUREMENTS OF THE RHONE GLACIER VELOCITY AND TOPOGRAPHY USING IN-SITU REAL-APERTURE RADAR

Satellite interferometry has been used extensively for remote sensing of glaciers with good success. There are however important limitations related to the fixed imaging geometry, moderate resolution, and typical repeat intervals of 3-6 weeks. Interferograms from satellite SAR data suffer from strong decorrelation due to a combination of large displacements and surface change. These considerations strongly motivated the development of a ground-based interferometric radar sensor with the capability to make high-resolution measurements of glacier topography and surface velocity within a single day.
We report the results of in-situ measurements of the Rhone Glacier in Switzerland using our newly developed real-aperture interferometric radar sensor. The Terrestrial Interferometric Radar was deployed overlooking the Rhone Glacier in September and October 2007 for a series of measurements to evaluate the feasibility of using a real-aperture interferometric radar for remote sensing of glaciers.
Comparisons of the phase between successive images acquired from the same viewpoint can be used to determine displacements on the order of a fraction of a wavelength. In the case of this instrument, operating at a wavelength of 17.4 mm, the measurement sensitivity is < 1mm.
Multiple images of the Rhone glacier where acquired over a 6 hour period. Interferometric post-processing of these data yielded LOS velocity estimates on the order of 4 mm/hr (35 m/year). These are in good agreement with velocities derived using photogrammetry. The interferograms have good correlation even over several hours. Decorrelation on the glacier surface is observed due to changes of the surface backscatter due to melting and increased for longer time intervals.
We also acquired simultaneous interferometric image pairs with a vertical baseline of 25 cm. We derived the surface topography from the interferometric phase by calculating the precise angle of the LOS relative to the baseline. Rapid simultaneous acquisition of the image pair eliminates interferometric decorrelation due to surface change and phase errors due to tropospheric variability and motion. Our topographic data are in good agreement with the DHM-25 topographic map and photogrammetric data.