Day 4

Detailed paper information

Back to list

Paper title Evaluation of the Effects of Ice Velocity Errors on an Inverse Model of the Filchner-Ronne Sector of Antarctica
Authors
  1. Michael Wolovick Alfred Wegner Institute Helmholtz Centre for Polar and Marine Science Speaker
  2. Lea-Sophie Höyns Alfred-Wegener-Institut - Helmholtz-Centre for Polar and Marine Research
  3. Thomas Kleiner Alfred Wegner Institute Helmholtz Centre for Polar and Marine Science
  4. Niklas Neckel Alfred Wegner Institute Helmholtz Centre for Polar and Marine Science
  5. Veit Helm Alfred-Wegener-Institut Helmholtz-Zentrum für Polar- und Meeresforschung
  6. Angelika Humbert Alfred Wegener Institute Helmholtz Center for Polar and Marine Research
Form of presentation Poster
Topics
  • C2. Digital Twins
    • C2.01 Towards a Digital Twin of the Earth - advances and challenges ahead
Abstract text Inverse models are a vital tool for inferring the state of ice sheets based on remote sensing data. Remotely sensed observations of ice surface velocity can be combined with a numerical model of ice flow to reconstruct the stress and deformation fields inside the ice and to infer the basal drag and/or englacial rheology. However, velocity products based on remote sensing contain both random and correlated errors, often including artifacts aligned with satellite orbits or particular latitude bands. Here, we use a higher-order inverse model within the Ice Sheet System Model (ISSM; Larour et al., 2012) to assimilate satellite observations of ice surface velocity in the Filchner-Ronne sector of Antarctica in order to infer basal drag underneath the grounded ice and englacial rheology in the floating shelf ice. We use multiple velocity products to constrain our model, including MEaSUREs_v2 (Rignot et al., 2017), an updated version of the MEaSUREs dataset that incorporates estimates from SAR interferometry (Mouginot et al., 2019), and a new mosaic for this sector that combines data from Sentinel-1, Landsat-8, and TerraSAR-X (Hofstede et al., 2021). For each velocity source, we perform an independent L-curve analysis to determine the optimal degree of spatial smoothing (regularization) needed to fit the observations without overfitting to noise. Additionally, we test the sensitivity of the inverted results to increased noise levels in the input data, using both random normally distributed noise and correlated noise constructed to resemble the satellite-orbit patches often found in ice velocity products. Using the L-curve analysis, we evaluate which remotely sensed velocity product permits the highest-resolution reconstruction of basal drag or englacial rheology. We find that correlated errors and artifacts in the velocity data produce corresponding artifacts in the inverse model results, particularly in the floating part where the inverted rheology estimate is highly sensitive to spatial gradients of the observed velocity field. The inversion for basal drag in the grounded ice displays less sensitivity to artifacts in the input data, because the drag inversion is less dependent on spatial gradients of the observed velocities. Minimizing the rheology artifacts in the floating shelf ice requires increased regularization of the inversion, thus reducing the spatial resolution of the inversion result. Because of the large spatial scale of the artifacts present in the velocity products, it is impossible to completely remove the corresponding artifacts in the inversion result without imposing such a degree of regularization that real structure (such as shear margins and rifts) is lost. By contrast, the inversion results are quite robust to uncorrelated errors in the input data. We suggest that future attempts to construct estimates of the ice surface velocity from remote sensing data should take care to remove correlated errors and “stripes” from their final product, and that inversion results for englacial rheology are particularly sensitive to artifacts that appear in the gradients of the observed velocity.

References:

Hofstede, C., Beyer, S., Corr, H., Eisen, O., Hattermann, T., Helm, V., Neckel, N., Smith, E. C., Steinhage, D., Zeising, O., and Humbert, A.: Evidence for a grounding line fan at the onset of a basal channel under the ice shelf of Support Force Glacier, Antarctica, revealed by reflection seismics, The Cryosphere, 15, 1517–1535, https://doi.org/10.5194/tc-15-1517-2021, 2021.

E. Larour, H. Seroussi, M. Morlighem, and E. Rignot (2012), Continental scale, high order, high spatial resolution, ice sheet modeling using the Ice Sheet System Model, J. Geophys. Res., 117, F01022, doi:10.1029/2011JF002140.

Mouginot, J., Rignot, E., & Scheuchl, B. (2019). Continent-wide, interferometric SAR phase, mapping of Antarctic ice velocity. Geophysical Research Letters, 46, 9710– 9718. https://doi.org/10.1029/2019GL083826

Rignot, E., J. Mouginot, and B. Scheuchl. 2011. Ice Flow of the Antarctic Ice Sheet, Science. 333. 1427-1430. https://doi.org/10.1126/science.1208336

Rignot, E., J. Mouginot, and B. Scheuchl. 2017. MEaSUREs InSAR-Based Antarctica Ice Velocity Map, Version 2. Boulder, Colorado USA. NASA National Snow and Ice Data Center Distributed Active Archive Center. doi: https://doi.org/10.5067/D7GK8F5J8M8R