the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 17 Jul 2025
Freshly calved icebergs from Sermeq Kujalleq in Kangia, Greenland: is their blue ice temperate?
Abstract. Blue ice on the freshly calved icebergs from Sermeq Kujalleq in Kangia, Greenland, is a striking feature which is currently barely understood. Its extent and properties provide important insights into the flow dynamics of this polar ice stream, since it occupies the bottom-most quarter of the total ice thickness, and it has been conjectured that the blue ice is temperate. Here, we document the phenomenon with ground-based time-lapse camera images, theodolite measurements, thermal imagery, and multi-spectral satellite imaging from Sentinel-2. The blue ice shows intriguing properties as its reflectance spectrum stands out from other types of ice and resembles that of water. The blue ice whitens under the action of solar radiation, potentially caused by the drainage of liquid water, which would indicate temperate conditions. Thermal imaging on the other hand refutes this interpretation, indicating that blue ice is colder than white ice when using a standard thermal emissivity. Freshly calved icebergs offer a unique opportunity to explore properties and structures (layering, folding) of ice which are formed at great depth within polar ice streams.
-
Notice on discussion status
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
-
Preprint
(9973 KB)
-
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(9973 KB) - Metadata XML
- BibTeX
- EndNote
- Final revised paper
Journal article(s) based on this preprint
Interactive discussion
Status: closed
-
RC1: 'Review by Stephen Warren', Stephen Warren, 05 Aug 2025
-
AC1: 'Reply on RC1', Antoine Zaninetti, 17 Mar 2026
Dear Stephen,
Thank you for your comments. We explored your hypothesis about clathrates hydrates and incorporated it in the article. There are some aspects that would need to be further demonstrated (lower overburden pressure in the final part of the ice stream, why sometimes the presence of white ice below the blue ice), but it is a really intersting thought. You can find attached a summary of the changes to your smaller comments.
Best regards,
Antoine Zaninetti
-
AC1: 'Reply on RC1', Antoine Zaninetti, 17 Mar 2026
-
RC2: 'Comment on egusphere-2025-2963', Anonymous Referee #2, 20 Aug 2025
Manuscript summary
Zaninetti et al. report newly collected field observations of thick blue layers in freshly calved icebergs from Sermeq Kujalleq, Greenland, collected in July 2023 and other unspecified dates in the autumn. This follows on the interesting analysis begun by Luthi et al. (2009), which was a popular correspondence paper that raised the interesting question that Zaninetti et al. advance here. The observations are interesting and well documented, but the manuscript's current conclusions are too wishy-washy. I believe that with the ingestion of the new hypothesis by Reviewer 1, which I find quite compelling, this manuscript would be significantly strengthened and would make a very nice contribution to the literature on Sermeq Kujalleq.
Major pointsI find myself largely in agreement with Reviewer 1, at least as far as my expertise goes (he knows much more about the optical properties of ice than I do). What I mean by my agreement is that I am largely satisfied by the observations and their documentation / presentation in figures -- these are nicely done! -- but I think that the inferences and conclusions laid out are too tentative, speculative, and don't explain the observations strongly enough. I believe that Reviewer 1's idea of clathrate release provides a better explanation and that the authors should scrutinize, adopt, and incorporate it. Many of my points here stem from implications and second-order effects of this new hypothesis that I would encourage the authors to address in a revision.
Line 83-85 - "blue ice is visible during almost all major calving events and appears to be independent of the position of the iceberg along the calving front, indicating that it exists across the entire width of the ice stream." If the clathrate hypothesis is correct, ice thickness of at least ~1600 m would be required along the entire calving front in order to support this statement / observation. This may not be the case, as the trough becomes shallower away from the centerline. The authors should present evidence from BedMachine (or an accepted ice thickness dataset of their choice) that this is indeed the case, or discuss the limits of the position along the calving front where blue ice should be present. A problematic deduction is that of Luthi et al. (2009): that the calving front was "only" 900 meters thick at the time. This may have changed as the front retreated into deeper water over the past 15 years -- but still, the blue ice was observed then, too. Could or should their methods for estimating local ice thickness be revisited or revised? Or perhaps the transit time from a location where H>1600 m to the 2009 calving front is sufficiently short (~days) that clathrates would not yet decompress?
Line 96 - "Temperature is not absolute but on all profiles, the blue ice is relatively colder than the white ice." This is in conflict with Reviewer 1’s hypothesis that all basal ice could/should be blue, regardless of whether it is temperate or cold. A solution would be if blue ice (with clathrates) has different thermal emissivity than white ice (without). I recommend that the authors please investigate this. I also ask that in the revision, they take care to distinguish between the observation (infrared emission) and the derived quantity (inferred temperature), since they differ by the emissivity which may be uncertain -- this uncertainty should be better discussed too.
Figure 5c shows that blue ice can re-emerge ~20 hours after calving if the ice fractures and exposes its interior. This observation is also problematic for Reviewer 1's hypothesis, as the iceberg is no longer under the pressure required to maintain clathrates. I have no good suggestions for this apparent contradiction, aside from the timescale (mentioned above) of clathrate release (which should be reported) and very weak speculation about some chemical interaction with air (N2 or O2) or mechanical wind action. The apparent contradiction should be discussed.
Line 162-163 - "Englacial liquid water seems to be a plausible explanation of a highly absorbing substance that drains out of the ice by gravity". A test of this hypothesis would be whether a whiteness gradient from top to bottom was observed during the period of reflectance change. It is not evident from the observations presented here, but I would like to know if this was in fact observed. Of course, the iceberg is tipping around in the fjord as it melts, which changes its orientation, so “down” is not always in exactly the same direction in the iceberg's reference frame.
Line 227 - "The vertical position of the blue ice is not necessarily at the bottom of the ice stream (Lüthi et al., 2009)." This would be inconsistent with Reviewer 1's hypothesis, unless temperate ice for some reason cannot form clathrates. Could the interstitial liquid water get in the way of the intra-granular clathrate? This should be investigated and commented on.
Minor pointsLine 19 - "these colors already exist within the ice stream." I can deduce that the authors mean that the ice stream is fast-moving and grounded so there is no possibility of marine ice accretion, but this should be better specified.
Line 38 - "From extrapolation of measured temperature profiles it was concluded that the fast flow is caused by
the enhanced deformation of a thick basal layer of temperate ice within the trough." This needs citation or rephrasing, because modeling studies in the past decade have shown that despite significant deformation (as found by Iken et al. 1993 and Funk et al. 1994), the fast flow is still accommodated mostly by sliding (Poinar et al. 2017, Shapero et al. 2016).Line 49 - "shaded" - Refer to this as "in shadow" instead.
Line 86 - "For icebergs that rise vertically out of the water, the upper part emerging out of the water disintegrates, leaving the lower portion presenting blue ice rising by buoyancy." I am not sure what this description of the icebergs means. Does it refer to tabular icebergs, which present in their "natural" orientation, rather than the vast majority of icebergs that flip immediately as they calve? I think I must misunderstand which icebergs are being discussed here; please clarify.
Line 120 - "long-term evolution" - a different term is recommended, as 12 days is not likely to be thought of as a long period in glaciology
Line 130 - What are the autumn observation dates? Also, the summer dates should also be specified; I have assumed that they are the July 2023 field dates, but this may not be the case.
Line 138 - the BIA are referred to here as "well studied", which is in contradiction to the single reference repeatedly used for them (Hui et al., 2014). Additional references would bolster this claim, or simply remove the descriptor "well studied".
Line 138 - "no reference spectra were found for temperate ice" - Pope & Rees (2014) studied broadspectrum albedo on temperate Icelandic glaciers
Line 148 - "It however appears in the mentioned literature" - Please cite it again, since I am not sure which specific references are meant if I want to look into this statement.
Line 150 - "surprisingly" - please remove this non-objective word
Line 154-158 - I also found this paragraph to be muddled and needing improvement. I'd add to Reviewer 1's comments that "opaque" and "reflecting" are not antonyms.
Line 193-194 - That the icebergs "rose from the water 30 to 60 seconds later in a second phase after the disintegration of the upper part" should be moved to the Results as they are direct observations. The generation of the icebergs should be described more thoroughly overall.
Line 200 - Within measurement uncertainty and additional uncertainly stemming from unknown emissivity, is the ice here in the cold core of B5 truly any colder than the white ice in B5?
Line 201 - The number 0.5 refers, I think, to the "halfway depth" of the ice stream. This should be better defined.
Reviewer 1 expects the ice to turn gray first and then white as the clathrates release. Why the gray phase, and was this transition observed? I recognize that the authors may not have an answer for this, as this is the Reviewer's idea (not theirs), but I am asking simply because I would like to know.
Perhaps a related question to the above: What do the authors hypothesize to be the origin of the grayish-green layers? This observation is reported, but its cause never explored (it only appears in Section 4.6, "Pathways for further research").
Figure 4 caption - Please fix the typo "in black line" and define "BIA" in the caption (it is defined in the text, but this was harder to find).
Figure 5 caption - "except that water was trapped during calving (no melt water)" - What does this mean? Is it implied that this is seawater, not water sourced from the glacier? Please clarify.
References
Pope and Rees, 2014. “Using in Situ Spectra to Explore Landsat Classification of Glacier Surfaces.” International Journal of Applied Earth Observation and Geoinformation 27: 42–52. doi:10.1016/j.jag.2013.08.007.
Shapero, Joughin, Poinar, Morlighem, and Gillet-Chaulet, 2016. “Basal Resistance for Three of the Largest Greenland Outlet Glaciers.” Journal of Geophysical Research 121 (January): 1–13. doi:10.1002/2015jf003643.
Citation: https://doi.org/10.5194/egusphere-2025-2963-RC2 -
AC2: 'Reply on RC2', Antoine Zaninetti, 17 Mar 2026
Dear reviewer,
Thank you for your constructive and intersting comments. We incorporated and discussed the clathrates hypothesis in the discussion, which sounds plausible.
You asked several questions about the clathrates, that shaped our discussion with pros and cons. Some questions remain unanswered due to the lack of data or general scientific knowledge. You can find attached a summary of the changes to your comments. The main changes are directly readable in the manuscript.
Best regards,
Antoine Zaninetti
-
AC2: 'Reply on RC2', Antoine Zaninetti, 17 Mar 2026
Interactive discussion
Status: closed
-
RC1: 'Review by Stephen Warren', Stephen Warren, 05 Aug 2025
This is an interesting paper that addresses two major questions:
(1) What is the temperature (and emissivity) of the blue ice?
(2) Why is the ice clear (and therefore blue)?
I have comments on both questions. “Major revision” is required for writing the discussion, conclusions, and abstract, but the figures probably do not need editing.
Major comments:
(1) What is the temperature (and emissivity) of the blue ice?
The analyses apparently assumed emissivity of 0.98 (Figure 3 legends to parts g and h). It could be that the ice is temperate (i.e. at 0°C), and that the low brightness-temperatures are caused by emissivity lower than 0.98, which is to be expected anyway for an oblique viewing angle. For example, at l=9mm, Figure 8e of Hori et al. (2006) shows the emissivity of bare ice to be 0.98 for viewing angles 0°-30°, but lower emissivity at more-oblique angles: 0.97 at 45°, 0.95 at 60°, and 0.82 at 75°.
(2) Why is the ice clear (and therefore blue)?
Clean glacier ice is normally white or gray because it contains numerous bubbles. But the blue ice in these icebergs is dark, indicating that it is mostly devoid of bubbles. This could be explained by melting of the glacier ice, allowing the bubbles to escape, then refreezing of the meltwater. But if the ice went through a melting process, what then caused the meltwater to refreeze?
The increase of albedo after 2, 3, and 12 days, shown in Figure 4a, indicates development of something that can cause refraction: either bubbles or cracks, or both.
I suggest a different explanation for the lack of bubbles in the dark-blue ice: The thickness of the ice stream is 2500 m, which means the pressure at the bed is 223 bars. At pressures in excess of 150 bars, the air-bubbles dissolve in ice to form clathrates (Figure 1 of Kuhs et al., 2000), resulting in clear ice, so the lowest 33% of the ice thickness would be at sufficiently high pressure to form clathrates. This is consistent with the estimate of Zaninetti et al. that only “the bottom-most quarter” of the ice thickness is dark blue. When the pressure is released, i.e. after the iceberg rises to the ocean surface and the blue ice is exposed to the atmosphere, the bubbles will re-form and the ice will turn gray and then white, as observed in Figure 5, consequently raising the albedo as shown in the dashed blue lines of Figure 4a after 2, 3, and 12 days.
Additional comments
Line 111, also the black curve in Figure 4a. The lake-reflectance of 0.01 for visible wavelengths is not representative of the lake’s albedo. This low value of reflectance is probably a directional reflectance under clear sky, where the measurement does not include the specular reflection (sunglint). The hemispheric albedo would be 0.07 for diffuse incidence (under cloud).
Figure 4 caption, line 3. Change “Lake reflectance” to “The reflectance of a land-based lake”.
Line 149. What is meant by “concave shape”? Do you mean concave-upward or concave-downward? And over what spectral region?
Lines 154-158. This paragraph about supraglacial lakes is muddled. The brilliant blue color of supraglacial lakes can be explained in the same way as the brilliant blue color of shallow water over white sand on Caribbean beaches: The submerged white sand, or the submerged white ice, has high reflectance for all visible wavelengths, but the overlying water acts as a filter, absorbing the red light but transmitting the blue. The absorption length (the mean-free-path of a photon before absorption; the reciprocal of the absorption coefficient) is ~2 m for red light but ~200 m for blue light.
Line 163. “highly absorbing substance”. Water is not “highly absorbing”. The absorption spectrum of liquid water across the visible wavelengths is similar to that of ice (Figure 1 of Dozier 1989; Figure 3 of Warren 2019). Both water and ice are nearly transparent for visible light.
Lines 208-209. Microwave emissivities are irrelevant for this work. Cite instead the thermal-infrared emissivities (Hori et al. 2006)
Grammar and spelling:
Line 21. Change “were” to “is”.
Line 23. Change boreholes to borehole.
Line 25. Change stream to streams.
Line 34. Change through to trough.
Line 69. Change “allowed to” to “allowed us to”.
Lines 70-71. Change “spaced by 240 m” to “spaced 240 m apart”.
Line 133. Change occured to occurred.
Line 216. Change “revert” to “invert”.
Line 295. Change Scambo to Scambos.
References:
Dozier J., 1989: Estimation of properties of alpine snow from Landsat Thematic Mapper. Adv. Space Res. 9(1), 207-215.
Hori M, Aoki T, Tanikawa T, Motoyoshi H, Hachikubo A, Sugiura K, Yasunari TJ, Eide H, Storvold R, Nakajima Y, Takahashi F., 2006: In-situ measured spectral direcitonal emissivity of snow and ice in the 8-14 mm atmospheric window. Rem. Sens. Environ. 100, 486-502.
Kuhs, W.F., A. Klapproth, and B. Chazallon, 2000: Chemical physics of air clathrate hydrates. In Physics of Ice Core Records (T. Hondon, Ed.), 373-392.
Warren, S.G., 2019: Optical properties of ice and snow. Phil. Trans. Royal Soc. A, 377, doi:10.1098/rsta.2018.016
Citation: https://doi.org/10.5194/egusphere-2025-2963-RC1 -
AC1: 'Reply on RC1', Antoine Zaninetti, 17 Mar 2026
Dear Stephen,
Thank you for your comments. We explored your hypothesis about clathrates hydrates and incorporated it in the article. There are some aspects that would need to be further demonstrated (lower overburden pressure in the final part of the ice stream, why sometimes the presence of white ice below the blue ice), but it is a really intersting thought. You can find attached a summary of the changes to your smaller comments.
Best regards,
Antoine Zaninetti
-
AC1: 'Reply on RC1', Antoine Zaninetti, 17 Mar 2026
-
RC2: 'Comment on egusphere-2025-2963', Anonymous Referee #2, 20 Aug 2025
Manuscript summary
Zaninetti et al. report newly collected field observations of thick blue layers in freshly calved icebergs from Sermeq Kujalleq, Greenland, collected in July 2023 and other unspecified dates in the autumn. This follows on the interesting analysis begun by Luthi et al. (2009), which was a popular correspondence paper that raised the interesting question that Zaninetti et al. advance here. The observations are interesting and well documented, but the manuscript's current conclusions are too wishy-washy. I believe that with the ingestion of the new hypothesis by Reviewer 1, which I find quite compelling, this manuscript would be significantly strengthened and would make a very nice contribution to the literature on Sermeq Kujalleq.
Major pointsI find myself largely in agreement with Reviewer 1, at least as far as my expertise goes (he knows much more about the optical properties of ice than I do). What I mean by my agreement is that I am largely satisfied by the observations and their documentation / presentation in figures -- these are nicely done! -- but I think that the inferences and conclusions laid out are too tentative, speculative, and don't explain the observations strongly enough. I believe that Reviewer 1's idea of clathrate release provides a better explanation and that the authors should scrutinize, adopt, and incorporate it. Many of my points here stem from implications and second-order effects of this new hypothesis that I would encourage the authors to address in a revision.
Line 83-85 - "blue ice is visible during almost all major calving events and appears to be independent of the position of the iceberg along the calving front, indicating that it exists across the entire width of the ice stream." If the clathrate hypothesis is correct, ice thickness of at least ~1600 m would be required along the entire calving front in order to support this statement / observation. This may not be the case, as the trough becomes shallower away from the centerline. The authors should present evidence from BedMachine (or an accepted ice thickness dataset of their choice) that this is indeed the case, or discuss the limits of the position along the calving front where blue ice should be present. A problematic deduction is that of Luthi et al. (2009): that the calving front was "only" 900 meters thick at the time. This may have changed as the front retreated into deeper water over the past 15 years -- but still, the blue ice was observed then, too. Could or should their methods for estimating local ice thickness be revisited or revised? Or perhaps the transit time from a location where H>1600 m to the 2009 calving front is sufficiently short (~days) that clathrates would not yet decompress?
Line 96 - "Temperature is not absolute but on all profiles, the blue ice is relatively colder than the white ice." This is in conflict with Reviewer 1’s hypothesis that all basal ice could/should be blue, regardless of whether it is temperate or cold. A solution would be if blue ice (with clathrates) has different thermal emissivity than white ice (without). I recommend that the authors please investigate this. I also ask that in the revision, they take care to distinguish between the observation (infrared emission) and the derived quantity (inferred temperature), since they differ by the emissivity which may be uncertain -- this uncertainty should be better discussed too.
Figure 5c shows that blue ice can re-emerge ~20 hours after calving if the ice fractures and exposes its interior. This observation is also problematic for Reviewer 1's hypothesis, as the iceberg is no longer under the pressure required to maintain clathrates. I have no good suggestions for this apparent contradiction, aside from the timescale (mentioned above) of clathrate release (which should be reported) and very weak speculation about some chemical interaction with air (N2 or O2) or mechanical wind action. The apparent contradiction should be discussed.
Line 162-163 - "Englacial liquid water seems to be a plausible explanation of a highly absorbing substance that drains out of the ice by gravity". A test of this hypothesis would be whether a whiteness gradient from top to bottom was observed during the period of reflectance change. It is not evident from the observations presented here, but I would like to know if this was in fact observed. Of course, the iceberg is tipping around in the fjord as it melts, which changes its orientation, so “down” is not always in exactly the same direction in the iceberg's reference frame.
Line 227 - "The vertical position of the blue ice is not necessarily at the bottom of the ice stream (Lüthi et al., 2009)." This would be inconsistent with Reviewer 1's hypothesis, unless temperate ice for some reason cannot form clathrates. Could the interstitial liquid water get in the way of the intra-granular clathrate? This should be investigated and commented on.
Minor pointsLine 19 - "these colors already exist within the ice stream." I can deduce that the authors mean that the ice stream is fast-moving and grounded so there is no possibility of marine ice accretion, but this should be better specified.
Line 38 - "From extrapolation of measured temperature profiles it was concluded that the fast flow is caused by
the enhanced deformation of a thick basal layer of temperate ice within the trough." This needs citation or rephrasing, because modeling studies in the past decade have shown that despite significant deformation (as found by Iken et al. 1993 and Funk et al. 1994), the fast flow is still accommodated mostly by sliding (Poinar et al. 2017, Shapero et al. 2016).Line 49 - "shaded" - Refer to this as "in shadow" instead.
Line 86 - "For icebergs that rise vertically out of the water, the upper part emerging out of the water disintegrates, leaving the lower portion presenting blue ice rising by buoyancy." I am not sure what this description of the icebergs means. Does it refer to tabular icebergs, which present in their "natural" orientation, rather than the vast majority of icebergs that flip immediately as they calve? I think I must misunderstand which icebergs are being discussed here; please clarify.
Line 120 - "long-term evolution" - a different term is recommended, as 12 days is not likely to be thought of as a long period in glaciology
Line 130 - What are the autumn observation dates? Also, the summer dates should also be specified; I have assumed that they are the July 2023 field dates, but this may not be the case.
Line 138 - the BIA are referred to here as "well studied", which is in contradiction to the single reference repeatedly used for them (Hui et al., 2014). Additional references would bolster this claim, or simply remove the descriptor "well studied".
Line 138 - "no reference spectra were found for temperate ice" - Pope & Rees (2014) studied broadspectrum albedo on temperate Icelandic glaciers
Line 148 - "It however appears in the mentioned literature" - Please cite it again, since I am not sure which specific references are meant if I want to look into this statement.
Line 150 - "surprisingly" - please remove this non-objective word
Line 154-158 - I also found this paragraph to be muddled and needing improvement. I'd add to Reviewer 1's comments that "opaque" and "reflecting" are not antonyms.
Line 193-194 - That the icebergs "rose from the water 30 to 60 seconds later in a second phase after the disintegration of the upper part" should be moved to the Results as they are direct observations. The generation of the icebergs should be described more thoroughly overall.
Line 200 - Within measurement uncertainty and additional uncertainly stemming from unknown emissivity, is the ice here in the cold core of B5 truly any colder than the white ice in B5?
Line 201 - The number 0.5 refers, I think, to the "halfway depth" of the ice stream. This should be better defined.
Reviewer 1 expects the ice to turn gray first and then white as the clathrates release. Why the gray phase, and was this transition observed? I recognize that the authors may not have an answer for this, as this is the Reviewer's idea (not theirs), but I am asking simply because I would like to know.
Perhaps a related question to the above: What do the authors hypothesize to be the origin of the grayish-green layers? This observation is reported, but its cause never explored (it only appears in Section 4.6, "Pathways for further research").
Figure 4 caption - Please fix the typo "in black line" and define "BIA" in the caption (it is defined in the text, but this was harder to find).
Figure 5 caption - "except that water was trapped during calving (no melt water)" - What does this mean? Is it implied that this is seawater, not water sourced from the glacier? Please clarify.
References
Pope and Rees, 2014. “Using in Situ Spectra to Explore Landsat Classification of Glacier Surfaces.” International Journal of Applied Earth Observation and Geoinformation 27: 42–52. doi:10.1016/j.jag.2013.08.007.
Shapero, Joughin, Poinar, Morlighem, and Gillet-Chaulet, 2016. “Basal Resistance for Three of the Largest Greenland Outlet Glaciers.” Journal of Geophysical Research 121 (January): 1–13. doi:10.1002/2015jf003643.
Citation: https://doi.org/10.5194/egusphere-2025-2963-RC2 -
AC2: 'Reply on RC2', Antoine Zaninetti, 17 Mar 2026
Dear reviewer,
Thank you for your constructive and intersting comments. We incorporated and discussed the clathrates hypothesis in the discussion, which sounds plausible.
You asked several questions about the clathrates, that shaped our discussion with pros and cons. Some questions remain unanswered due to the lack of data or general scientific knowledge. You can find attached a summary of the changes to your comments. The main changes are directly readable in the manuscript.
Best regards,
Antoine Zaninetti
-
AC2: 'Reply on RC2', Antoine Zaninetti, 17 Mar 2026
Peer review completion
Journal article(s) based on this preprint
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 7,774 | 2,282 | 207 | 10,263 | 371 | 448 |
- HTML: 7,774
- PDF: 2,282
- XML: 207
- Total: 10,263
- BibTeX: 371
- EndNote: 448
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Antoine Paul Zaninetti
Martin P. Lüthi
Adrien Justin Wehrlé
Janneke van Ginkel
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(9973 KB) - Metadata XML
This is an interesting paper that addresses two major questions:
(1) What is the temperature (and emissivity) of the blue ice?
(2) Why is the ice clear (and therefore blue)?
I have comments on both questions. “Major revision” is required for writing the discussion, conclusions, and abstract, but the figures probably do not need editing.
Major comments:
(1) What is the temperature (and emissivity) of the blue ice?
The analyses apparently assumed emissivity of 0.98 (Figure 3 legends to parts g and h). It could be that the ice is temperate (i.e. at 0°C), and that the low brightness-temperatures are caused by emissivity lower than 0.98, which is to be expected anyway for an oblique viewing angle. For example, at l=9mm, Figure 8e of Hori et al. (2006) shows the emissivity of bare ice to be 0.98 for viewing angles 0°-30°, but lower emissivity at more-oblique angles: 0.97 at 45°, 0.95 at 60°, and 0.82 at 75°.
(2) Why is the ice clear (and therefore blue)?
Clean glacier ice is normally white or gray because it contains numerous bubbles. But the blue ice in these icebergs is dark, indicating that it is mostly devoid of bubbles. This could be explained by melting of the glacier ice, allowing the bubbles to escape, then refreezing of the meltwater. But if the ice went through a melting process, what then caused the meltwater to refreeze?
The increase of albedo after 2, 3, and 12 days, shown in Figure 4a, indicates development of something that can cause refraction: either bubbles or cracks, or both.
I suggest a different explanation for the lack of bubbles in the dark-blue ice: The thickness of the ice stream is 2500 m, which means the pressure at the bed is 223 bars. At pressures in excess of 150 bars, the air-bubbles dissolve in ice to form clathrates (Figure 1 of Kuhs et al., 2000), resulting in clear ice, so the lowest 33% of the ice thickness would be at sufficiently high pressure to form clathrates. This is consistent with the estimate of Zaninetti et al. that only “the bottom-most quarter” of the ice thickness is dark blue. When the pressure is released, i.e. after the iceberg rises to the ocean surface and the blue ice is exposed to the atmosphere, the bubbles will re-form and the ice will turn gray and then white, as observed in Figure 5, consequently raising the albedo as shown in the dashed blue lines of Figure 4a after 2, 3, and 12 days.
Additional comments
Line 111, also the black curve in Figure 4a. The lake-reflectance of 0.01 for visible wavelengths is not representative of the lake’s albedo. This low value of reflectance is probably a directional reflectance under clear sky, where the measurement does not include the specular reflection (sunglint). The hemispheric albedo would be 0.07 for diffuse incidence (under cloud).
Figure 4 caption, line 3. Change “Lake reflectance” to “The reflectance of a land-based lake”.
Line 149. What is meant by “concave shape”? Do you mean concave-upward or concave-downward? And over what spectral region?
Lines 154-158. This paragraph about supraglacial lakes is muddled. The brilliant blue color of supraglacial lakes can be explained in the same way as the brilliant blue color of shallow water over white sand on Caribbean beaches: The submerged white sand, or the submerged white ice, has high reflectance for all visible wavelengths, but the overlying water acts as a filter, absorbing the red light but transmitting the blue. The absorption length (the mean-free-path of a photon before absorption; the reciprocal of the absorption coefficient) is ~2 m for red light but ~200 m for blue light.
Line 163. “highly absorbing substance”. Water is not “highly absorbing”. The absorption spectrum of liquid water across the visible wavelengths is similar to that of ice (Figure 1 of Dozier 1989; Figure 3 of Warren 2019). Both water and ice are nearly transparent for visible light.
Lines 208-209. Microwave emissivities are irrelevant for this work. Cite instead the thermal-infrared emissivities (Hori et al. 2006)
Grammar and spelling:
Line 21. Change “were” to “is”.
Line 23. Change boreholes to borehole.
Line 25. Change stream to streams.
Line 34. Change through to trough.
Line 69. Change “allowed to” to “allowed us to”.
Lines 70-71. Change “spaced by 240 m” to “spaced 240 m apart”.
Line 133. Change occured to occurred.
Line 216. Change “revert” to “invert”.
Line 295. Change Scambo to Scambos.
References:
Dozier J., 1989: Estimation of properties of alpine snow from Landsat Thematic Mapper. Adv. Space Res. 9(1), 207-215.
Hori M, Aoki T, Tanikawa T, Motoyoshi H, Hachikubo A, Sugiura K, Yasunari TJ, Eide H, Storvold R, Nakajima Y, Takahashi F., 2006: In-situ measured spectral direcitonal emissivity of snow and ice in the 8-14 mm atmospheric window. Rem. Sens. Environ. 100, 486-502.
Kuhs, W.F., A. Klapproth, and B. Chazallon, 2000: Chemical physics of air clathrate hydrates. In Physics of Ice Core Records (T. Hondon, Ed.), 373-392.
Warren, S.G., 2019: Optical properties of ice and snow. Phil. Trans. Royal Soc. A, 377, doi:10.1098/rsta.2018.016