Spatially and Temporally Dense Grounding Lines from Bayesian Inversion of Sentinel-1 Data

Grounding lines are flux gates through which ice discharges into the ocean. Their position reflects ice sheet stability, retreating landward or advancing seaward in response to changes in melting and accumulation, while also exhibiting short-term motion driven by tidal flexure of floating ice. Grounding lines derived from Differential Interferometric SAR (DInSAR) phase are generally regarded as the most accurate [1]. However, most existing datasets lack formal uncertainty estimates, even though grounding line errors directly propagate into ice discharge calculations and can substantially bias estimates of ice mass loss and sea level rise [1]. A further limitation is that DInSAR grounding lines are derived from interferograms combining three or four SAR acquisitions, such that each estimated position represents a superposition of multiple tidal states, complicating the attribution of observed displacements to specific tidal forcing.

We have developed a framework to obtain grounding line positions together with Bayesian estimates of positional uncertainty. Based on this, we generated a dense time series of grounding lines for the Larsen C Ice Shelf spanning 2019–2021, derived from Sentinel-1 line-of-sight (LOS) offsets at a temporal sampling of 6 days. The LOS offsets are part of the operational processing pipeline used by ENVEO IT to produce monthly and annual Sentinel-1 ice-velocity maps, and were computed by tracking features between consecutive SAR backscatter images [2], [3]. Grounding line positions were estimated by fitting the LOS offsets to a one-dimensional elastic beam model [4] and performing Bayesian inversion using the cross entropy based importance sampling for Bayesian updating (CEBU) algorithm [5], which allows the incorporation of external datasets as priors on model parameters. Additionally, the estimated error of the range offsets were explicitly accounted for in the inversion. The resulting dataset provides a dense time series of grounding lines which have a mean distance of 348.07 m from Sentinel-1 DInSAR grounding lines. Because the dataset is derived from SAR backscatter rather than interferometric phase, it is robust to coherence loss and can be used to fill gaps in DInSAR grounding line products over fast-flowing outlet glaciers and ice streams.

References

[1] E. Rignot, J. Mouginot, and B. Scheuchl, “Antarctic grounding line mapping from differential satellite radar interferometry: GROUNDING LINE OF ANTARCTICA,” Geophysical Research Letters, vol. 38, no. 10, 2011

[2] T. Nagler, H. Rott, M. Hetzenecker, J. Wuite, and P. Potin, “The sentinel-1 mission: New opportunities for ice sheet observations,” Remote Sensing, vol. 7, no. 7, pp. 9371–9389, 2015.

[3] J. Wuite, T. Nagler, M. Hetzenecker, and H. Rott, “Ten years of polar ice velocity mapping using Copernicus Sentinel-1,” Remote Sensing of Environment, vol. 332, p. 115 092, Jan. 2026

[4] G. Holdsworth, “Flexure of a Floating Ice Tongue,” Journal of Glaciology, vol. 8, no. 54, pp. 385–397, 1969, 1727-5652

[5] M. Engel, O. Kanjilal, I. Papaioannou, and D. Straub, “Bayesian updating and marginal likelihood estimation by cross entropy based importance sampling,” Journal of Computational Physics, vol. 473, p. 111 746, Jan. 2023