Day 4

Detailed paper information

Back to list

Paper title Extent of ice flow changes in the North East Greenland Ice Stream as seen by Sentinel-1 DInSAR time series
  1. Jonas Kvist Andersen DTU Space - Technical University of Denmark Speaker
  2. Jeremie Mouginot Univ. Grenoble Alpes, CNRS, Institut des Geosciences de l'Environnement
  3. John Peter Merryman Boncori DTU Space - Technical University of Denmark
  4. Anders Kusk DTU Space - Technical University of Denmark
Form of presentation Poster
  • A9. Polar Science and Cryosphere
    • A9.04 Mass Balance of the Cryosphere
Abstract text For several decades, Synthetic Aperture Radar (SAR) satellites have been applied in the measurement of the velocity of glaciers and ice sheets. Compared to earlier missions, the Sentinel-1 satellites, with a wide-swath acquisition mode and a 6-day repeat-pass period, provide a polar data archive of unprecedented size, allowing for frequent revisit of outlet glaciers and the interior Greenland ice sheet. Amplitude-based tracking methods are routinely applied to generate average ice velocity measurements on time scales ranging from a month to multiple years. On shorter time scales, noise levels in tracking-based measurements approach tens of m/y [1, 2], which is close to the signal level in the ice sheet interior and upstream parts of glaciers. Conversely, Differential SAR Interferometry (DInSAR), which is based on the radar phase signal, allows for velocity measurements with a significantly lower noise level ( < 0.5 m/y [3]) and higher resolution. Consequently, averaging of multiple acquisitions is generally not necessary to achieve measurements of high accuracy, even in slow-moving regions, and hence high quality velocity measurements can be made every six days.

A limitation of DInSAR is that it is only applicable in areas where interferometric coherence is retained, meaning that the very fast-flowing parts of outlet glaciers cannot be measured, due to phase aliasing. Hence, an obvious synergy exists between the tracking- and phase-based methods, which has been exploited for past SAR missions [1, 2]. For Sentinel-1, however, DInSAR has not been routinely applied in the retrieval of ice velocity, owing to additional challenges caused by a coupling between the differential phase and azimuth motion introduced by the TOPS acquisition mode. Recently, a solution to these challenges has been proposed [3, 4], unlocking the possibility for an improved exploitation of the Sentinel-1 archive.

The Northeast Greenland Ice Stream (NEGIS) is the only major dynamic feature of Greenland that extends continuously into the interior of the ice sheet near Greenland’s summit. Zachariae Isstrøm, Nioghalvfjerdsfjorden glacier and Storstrømmen, which form the NEGIS, drain an area representing more than 16% of the Greenland ice sheet. While Nioghalvfjerdsfjorden and Storstrømmen are still close to mass balance, Zachariae Isstrøm has begun a rapid retreat after detaching from a stabilizing sill in the late 1990s. Since 1999, the glacier flow has almost doubled and its acceleration has increased significantly after 2012, resulting in significant mass loss of this sector of Greenland [5]. Destabilization of this marine sector could increase sea level rise from the Greenland ice sheet for decades to come. While these changes in ice mass and motion are well documented near the ice margin, it remains to be established how the interior of the ice sheet responds to the change in stress balance that occurs at its margin. In other words, the extent to which multi-year and seasonal changes in dynamics due to variations in the calving front position propagate upstream of the glaciers is still unclear.

In this work, we apply Sentinel-1 DInSAR to generate a long, densely sampled time series of ice velocity measurements for the NEGIS. All available Sentinel-1 acquisitions are used, meaning that the temporal sampling is 6 days (12 days prior to the launch of Sentinel-1B) and the spatial sampling is 50x50 m. The goal is to investigate any long- and/or short-term changes in velocity as well as seasonal effects on the NEGIS. Similar studies have previously been carried out with tracking-based methods, typically focusing on the downstream parts of glaciers, where changes in velocity exceed the amplitude-tracking noise levels. In this study, we focus on velocity changes in the slower-moving upstream parts, where the higher accuracy and spatial/temporal resolution of DInSAR allows for significantly improved results.

Finally, we discuss the benefits and challenges of SAR interferometry compared to tracking methods in monitoring dynamical changes and conclude on the amplitude and extent of the current flow acceleration of the NEGIS due to the recent retreat of Zachariae Isstrøm.

[1] I. Joughin, B. E. Smith, and I. M. Howat, “A complete map of Greenland ice velocity derived from satellite data collected over 20 years,” Journal of Glaciology, vol. 64, no. 243, pp. 1–11, (2018)
[2] J. Mouginot, E. Rignot, and B. Scheuchl, “Continent-wide, interferometric SAR phase, mapping of Antarctic ice velocity,” Geophysical Research Letters, vol. 46, pp. 9710–9718, (2019)
[3] J. Andersen, A. Kusk, J. Boncori, C. Hvidberg, and A. Grinsted, “Improved ice velocity measurements with Sentinel-1 TOPS interferometry,” Remote Sensing, vol. 12, no. 12, (2020)
[4] A. Kusk, J. K. Andersen, and J. P. M. Boncori, “Burst overlap coregistration for Sentinel-1 TOPS DInSAR ice velocity measurements,” IEEE Geoscience and Remote Sensing Letters, (2021)
[5] J. Mouginot, E. Rignot, B. Scheuchl, I. Fenty, A. Khazendar, M. Morlighem, A. Buzzi, and J. Paden, "Fast retreat of Zachariæ Isstrøm, northeast Greenland", Science, vol. 350, no. 6266, (2015)