EGU2020-1173
https://doi.org/10.5194/egusphere-egu2020-1173
EGU General Assembly 2020
© Author(s) 2020. This work is distributed under
the Creative Commons Attribution 4.0 License.

Complex basal conditions influence flow at the onset of the North East Greenland Ice Stream

Steven Franke1, Daniela Jansen1, John Paden2, and Olaf Eisen1,3
Steven Franke et al.
  • 1Alfred-Wegener-Intitut, Helmholz-Zentrum für Polar und Meeresforschung, Glaciology, Bremerhaven, Germany
  • 2Center for Remote Sensing of Ice Sheets (CReSIS), University of Kansas, Lawrence, KS, USA
  • 3Department of Geosciences, University of Bremen, Bremen, Germany

The onset and high upstream ice surface velocities of the North East Greenland Ice Stream (NEGIS) are not yet well reproducible in ice sheet models. A major uncertainty remains the understanding of basal sliding and a parameterization of basal conditions. In this study, we assess the slow-flowing part of the NEGIS in a systematic analysis of the basal conditions and investigate the increased ice flow. We analyze the spectral basal roughness in correlation with basal return power from an airborne radar survey with AWIs ultra-wideband radar system in 2018 and compare our results with current ice flow geometry and ice surface flow. We observe a roughness anisotropy where the ice stream widens, indicating a change from a smooth and soft bed to a harder bedrock as well as the evolution of elongated subglacial landforms. In addition, at the upstream part of the NEGIS we find a clear zoning of the bedrock return power, indicating an increased water content at the base of the ice stream. At the downstream part, we observe an increased bedrock return power throughout the entire width of the ice stream and outside its margins, indicating enhanced melting and the distribution of basal water beyond the shear zones.

How to cite: Franke, S., Jansen, D., Paden, J., and Eisen, O.: Complex basal conditions influence flow at the onset of the North East Greenland Ice Stream, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-1173, https://doi.org/10.5194/egusphere-egu2020-1173, 2019

Comments on the presentation

AC: Author Comment | CC: Community Comment | Report abuse

Presentation version 1 – uploaded on 16 Apr 2020
  • CC1: Comment on EGU2020-1173, Rebecca Schlegel, 06 May 2020

    Hi Steven,

    can you give a quick introduction on how you measure the roughness? Is that frequency based? And do you need several receivers and sources for that method (e.g. could I use this method for ground-based radar as well)?

    Also whats the scale of the roughness, are we talking about meters or cm?

    Really interesting, I think we do very similar thing, only on different datasets!

    Cheers,

    Rebecca

  • AC1: Comment on EGU2020-1173, Steven Franke, 07 May 2020

    Hey Rebecca,

    sure : )

    I used two approaches for the roughness:

    1. a spectral roughness approach with two parameters that describe the dominance of a certain vertical amplitude or horizontal wavelength. As input, I use just the bed picks. It is very nicely explained in section 3.1. by Gudlaugsson et. al., 2013 (https://doi.org/10.1002/2013JF002714). My data set is quite large and so I used a Window of ~ 2km, which corresponds to 128 data points with a distance of ~ 15m. This is sort of the resolution. But your resolution will be much higher in your data set and you can also use any number of 2**n data points, which is larger than 5 in your analysis window. There are very nice examples in Li et. al. 2010 (Figure 3) (https://doi.org/10.3189/002214310794457326) to see what the vertical and horizontal roughness parameter describe. And most importantly, this approach describes roughness in along-track direction, so parallel to the flight trajectory.
    2. Then I used also the abruptness of the waveform as an indication for small-scale roughness and here it is important how the energy of the bed reflection is distributed in the twt domain. The idea is that if the reflected energy is mostly scattered on a rough reflector it will be recorded distributed over a larger time window. A specular reflection would result in a very narrow signal. It is very nicely explained in section 2.3.1. in Cooper et. al. 2019 (https://www.the-cryosphere.net/13/3093/2019/) and the references therein. They also do much more in terms of waveform and roughness analysis as I did...
      And here I think that when side reflections appear as layover in the nadir bed reflection or are difficult to distinguish, they will mostly come from across-track because the fk-migration will minimize the effect from along-track. It also depends on the beam geometry, where you would expect off-nadir returns. That's why in rough terrain, this method would describe roughness in cross-track. The resolution then depends on the frequency and the maximum length of the window you set that you consider as bed reflection.

    I think for your data it would be a nice asset, especially when you have a lot of additional information on the subglacial environment.

    And yes, let's stay in touch : )
    See you (typing) tomorrow.

    Steven