the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 17 Jun 2025
Iceberg influence on snow distribution and slush formation on Antarctic landfast sea ice from airborne multi-sensor observations
Abstract. Antarctic landfast sea ice fringes much of the coast of Antarctica and plays an important role for coastal ice-ocean-atmosphere interaction and ice shelf stability, as well as for the sea ice associated ecosystem. It is often characterized by embedded icebergs, which influence wind-driven snow distribution and properties. Using high-resolution data from an airborne multi-sensor survey over landfast sea ice in Atka Bay, Dronning Maud Land, in December 2022, we investigate the characteristics of extensive snow drifts around icebergs and their impact on flooding. An airborne quad-polarized, ultra-wideband microwave (UWBM) snow radar and laser scanner reveal persistent snow accumulation patterns around icebergs, with thick snow drifts on the windward side of icebergs, elongated lateral snow drifts parallel to the prevailing wind direction along their sides, and virtually snow-free regions with rough ice surfaces in their lee. The mass of the thick wind-facing and lateral snow drifts pushes the sea ice locally below sea level leading to flooding and slush formation at the base of the snow drifts. These heterogeneous snow-water-sea-ice interfaces cause increased cross-polarized backscatter due to depolarization in the UWBM radar returns, providing a means for slush detection by airborne radar surveys. Presence of slush is confirmed by ground-based electromagnetic induction sounding data as well as with in situ measurements. Our study documents the significant influence of icebergs on snow thickness variability and redistribution over landfast sea ice and for slush formation. Moreover, it demonstrates that the snow in the lee of icebergs is thin, resulting in high radar backscatter in SAR imagery. These insights improve our understanding of wind-driven snow distribution and its impact on flooding on iceberg-laden landfast sea ice, contributing to better assessments of snow transport, sea ice mass balance, and climate modeling around Antarctica.
Competing interests: Christian Haas is a member of the editorial board of The Cryosphere. The authors declare no other competing interests.
Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.- Preprint
(12323 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2025-2657', John Yackel, 14 Jul 2025
Dear Authors and Handling Editor,
Iceberg influence on snow distribution and slush formation on Antarctic landfast sea ice from airborne multi-sensor observations by Steven Franke and co-authors presents a highly novel multi-sensor remote sensing assessment of of iceberg infiltrated snow-covered landfast seasonal sea ice in Atka Bay, Antarctica focused on early December 2022 during the ANTSI campaign. The datasets consist of quad-polarized, ultra-wideband microwave (UWBM) radar from CReSiS U Kansas, airborne laser scanner (ALS), the Modular Airborne Camera System (MACS), four Global Navigation Satellite System (GNSS) antennas, TanDEM-X band SAR imagery, along with coincident ground-based measurements including electromagnetic (EM) induction sounding (GEM2) and in-situ snow depth and sea ice drilling conducted reasonably close in time to the remote sensing data.
The research approach and its datasets are, in my opinion, highly novel and unique. The manuscript is extremely well written and organized and includes some of the most exquisitely constructed illustrations I have seen in a long time. Figure 3 is one such example. This manuscript makes a strong contribution towards improved understanding of snow processes on Antarctic sea ice including the interpretation and use of FMCW and X-band SAR and their polarization capability for snow depth estimation. I recommend publication subject to minor revisions and addressing my questions below.
General Comment:
1) I found the Introduction written oddly in the sense that results/conclusions are alluded to on several occasions (L44-47; L67-72) without having read an objectives statement. I strongly recommend that the Introduction include explicitly written and tractable objectives statements and also remove the suggestive language as to what the results and conclusions of the analysis will show.
2) The Venturi effect or similar fluid dynamic principles appear to be operating here. I suggest the authors research and possibly mention this principal and relate the effect to blowing snow around obstacles such as icebergs and discuss whether or not they expect the wind speed to increase leeward of the icebergs further enhancing wind scouring to keep the snow cover thin.
3) The easterly wind direction is mentioned several times as the predominant wind direction. Can a wind rose be provided from the nearest weather station to support the Klöwer et al., 2013 study? Undoubtedly there are winds from other directions which can often produce secondary drifting patterns on the snowscape.
Minor Comments:
L256-258; L383-384; L420-422. While I generally agree with the statement that high backscatter snow covered sea ice corresponds to larger topographic roughness, one has to be careful in entirely associating this high backscatter with surface roughness (even though your surface roughness metric from the ALS DEM data suggest as much). For example, in the attached supplement I have uploaded, there is a small iceberg (highlighted in red box) which has high backscatter in the lee of the iceberg but does not show the high roughness in the center region of the lee (other than the lateral side edges as described by the authors). So, it apparently does not occur in all cases. In my opinion, it is equally likely that this thin snow region can permit the warmer air temperatures to produce higher basal snow layer temperature and brine volume, altering dielectrics and increase volume scattering (as you allude to in L279; L341-350 and elsewhere). In other words, it could be MORE than just surface roughness, especially for your Type 2 reflections. This process is nicely described in https://ieeexplore.ieee.org/abstract/document/9000883
Table 1 - is it possible to provide AFIN drill site labels for Figure 1 circles?
L335 ... typo 'single'
L358 .. while snow-ice formation horizon is a likely candidate, a brine-wetted snow snow layer and its effect on dielectric properties and scattering, owing to the warm air and snow temperatures, is equally likely.
L369 .. typo ... space needed
-
RC2: 'Comment on egusphere-2025-2657', Anonymous Referee #2, 08 Sep 2025
The manuscript 'Iceberg influence on snow distribution and slush formation on Antarctic landfast sea ice from airborne multi-sensor observations' by Franke et al. illustrates how grounded ice bergs on landfast Antarctic sea ice affects snow redistribution on the windward side of the icebergs that induces thick snow drift accummulation and flooding. This is observed from a multi-platform and scale campaign on Atka Bay from December 2022 using UWBM Radar, ALS, NIR imagery, ground-based validation and imaging SAR. The paper is well-written, however, the theoretical 'treatment' and 'handling' of radar scattering mechanisms (mostly speculated and neglectance of double-bounce scattering), in combination with assumption of snow only as dry and flooded snow, makes the paper slightly speculative and needs considerable improvement. Therefore, I recommend major revisions for this round. My comments are general for this round and do not deal with specific line-by-line comments for now. The paper is structually well-written with outstanding depiction and quality of figures that may require some change once my feedback is incorporated or defended.
General Comments:
1) My major concern is interrelated. a) The paper assumes radar reflections originating 'either the transition from snow to ice or the transition from dry snow to flooded snow, which could be wet snow / slush or refrozen, salty snow ice.' Now our issue is this. When the thick snow on thinner ice induces negative freeboard, yes, slush and refrozen snow-ice forms. But, what the authors missed (surprising) is that slush formation during water percolation into the snow also causes snow layers above the slush to be highly saline and completely brine-wetted. This is not 'dry-snow' especially at air and snow temperatures between -4 and -1C. The authors in Line 220 mention 'we specifically define our snow column thickness as dry-snow thickness. The dry-snow depth represents either the interface between snow and ice or the interface between dry snow and flooded snow (i.e., slush).' This treatment of 'snow column' neglects this intermediate but highly scattering layer of this brine-wetted snow volume, highly sensitive to radar waves, even before the scattering originates from the slush layers below. This needs to be accounted for and addressed.
b) My second concern is tied to my first concern above. The paper's treatment of scattering mechanisms depicted in Figure 3, sections 4.2 and 5.2, figure 5 is very speculative and vague. First of all, thick snow on Antarctic sea ice, exhibiting dirunal air and snow temperature changes during the spring season leads to constant microstructural variations vertically, with formation of melt-refreeze layers, snow grain metamorphism and most importantly (in this study context), changes in snow brine volume from both the brine-wetted layers (omitted in the study) and slush layers underneath. This significantly modifies the snow/slush dielectrics leading to ambigious penetration of radar waves. Next, the paper uses reflection and scattering side by side which needs to be corrected. Figure 5 for example uses snow reflection from surface, volume and interfaces, but they are scattering processes than reflection. Next, with an iceberg sitting grounded on sea ice, double bounce scattering is the major contributor from SAR imagery which also needs to be addressed. Next, the theoretical interpretation of scattering mechanisms needs some modeling evidence to be conclusive than speculative. But I also understand there are no in situ samples of snow and slush from the site (or do you?). I understand the logistic difficulties to collect samples from close to an iceberg due to risk of its tipping. But I feel, irrespective of that, there are slush studies from past that could be used to 'less speculatively' interpret the role of snow and slush on radar signatures. I also strongly suggest to remove 'dry snow' from the paper as the temperatures are well above the dry snow thresholds.
c) Assumption of snow densities: The paper assumes an average snow density of 0.38 ± 0.06gcm-3. Well, with such a flooded snow volume, the assumption of 0.38 is highly underestimated. Is it possible to conduct a quick sensitivity analysis to use both dry snow density and slush density from literature to average them and redo the snow thickness calculation (provided you consider slush also as a part of the snow volume)? Also, Figure 4, the classified imagery is also not interpretive unless value ranges are shown. For example, SAR backscatter showing 'low' and 'high' values can be of any range correct? unless we show it. Also, instead of classifying scattering mechanisms can Types 1 to 3, I suggest to use scattering from geophysical interfaces/volume (e.g. air/snow, within-snow, slush etc etc), as using Types makes it difficult for a future author to use them as a proper terminology. Makes sense?
These are my major comments for now. Like I said, the paper is novel but needs considerable changes with respect to my concerns. I am sure the manuscript can be considerably improved if you think about these changes (or defend).
Citation: https://doi.org/10.5194/egusphere-2025-2657-RC2
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 363 | 48 | 11 | 422 | 13 | 28 |
- HTML: 363
- PDF: 48
- XML: 11
- Total: 422
- BibTeX: 13
- EndNote: 28
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1