Assessing the TanDEM-X elevation bias due to SAR signal penetration for glacier mass balance measurements

The elevation bias due to signal penetration in bistatic InSAR DEMs is recognized as a main error source together with co-registration for estimating glacier mass balance with the DEM differencing method. For TanDEM-X DEMs, the elevation processed from X-band (9.65 GHz) SAR data can lie up to 4-8m lower than the actual snow/ice surface in alpine accumulation areas [1]. However, this bias can often be mitigated by differencing TanDEM-X acquisitions from the same season with unchanged SAR geometry, reducing penetration differences between DEMs. The relative importance of SAR signal penetration for accurate mass balance measurements also reduces with the length of the observation period.

Notably, methods have been developed to correct for SAR signal penetration bias, including estimating volumetric coherence and inverting it [2,3]. However, correction methods have rarely been tested and validated across entire TanDEM-X scenes with coincident ground truth measurements of the actual ice surface. [4] calculated signal penetration based on inversion of volumetric coherence on Union Glacier, Antarctica and validated the results against the optical REMA DEM mosaic over temporally stable surfaces.

A recent study on Aletsch Glacier has observed the elevation bias due to signal penetration in a time stamped TanDEM-X DEM by comparing it to a coincident DEM acquisition from Pléiades optical imagery [1]. Moreover, during an inter-comparison experiment on glacier elevation changes, airborne lidar validation DEMs were produced for Aletsch Glacier enabling a comparison of volumetric changes with TanDEM-X measurements [5].

We use these results to analyse the circumstances under which a signal penetration correction layer associated to the individual processed TanDEM-X DEMs can be used to generate bistatic X-band DEMs that reflect the actual ice/snow surface. We will assess the impact of a signal penetration correction on mass balance measurements similar to [6].

References

[1] Bannwart, Jacqueline, Livia Piermattei, Inés Dussaillant, Lukas Krieger, Dana Floricioiu, Etienne Berthier, Claudia Roeoesli, Horst Machguth, and Michael Zemp. 2024. “Elevation Bias Due to Penetration of Spaceborne Radar Signal on Grosser Aletschgletscher, Switzerland.” Journal of Glaciology, April, 1–15. https://doi.org/10.1017/jog.2024.37.

[2] Weber Hoen, E., and H.A. Zebker. 2000. “Penetration Depths Inferred from Interferometric Volume Decorrelation Observed over the Greenland Ice Sheet.” IEEE Transactions on Geoscience and Remote Sensing 38 (6): 2571–83. https://doi.org/10.1109/36.885204.

[3] Dall, Jørgen. 2007. “InSAR Elevation Bias Caused by Penetration Into Uniform Volumes.” IEEE Transactions on Geoscience and Remote Sensing 45 (7): 2319–24. https://doi.org/10.1109/TGRS.2007.896613.

[4] Rott, Helmut, Stefan Scheiblauer, Jan Wuite, Lukas Krieger, Dana Floricioiu, Paola Rizzoli, Ludivine Libert, and Thomas Nagler. 2021. “Penetration of Interferometric Radar Signals in Antarctic Snow.” The Cryosphere 15 (9): 4399–4419. https://doi.org/10.5194/tc-15-4399-2021.

[5] Piermattei, Livia, Michael Zemp, Christian Sommer, Fanny Brun, Matthias H. Braun, Liss M. Andreassen, Joaquín M. C. Belart, et al. 2024. “Observing Glacier Elevation Changes from Spaceborne Optical and Radar Sensors – an Inter-Comparison Experiment Using ASTER and TanDEM-X Data.” The Cryosphere 18 (7): 3195–3230. https://doi.org/10.5194/tc-18-3195-2024.

[6] Abdullahi, Sahra, David Burgess, Birgit Wessel, Luke Copland, and Achim Roth. 2023. “Quantifying the Impact of X-Band InSAR Penetration Bias on Elevation Change and Mass Balance Estimation.” Annals of Glaciology 64 (92): 396–410. https://doi.org/10.1017/aog.2024.7.