Comments on: Evidence suggesting frazil ice crystal formation at the front of Hisinger Glacier in Dickson Fjord, Northeast Greenland by Fleur Juliëtte Rooijakkers, Ebbe Poulsen, Eugenio Ruiz-Castillo, and Søren Rysgaard

This manuscript examines circulation and meltwater very near the terminal face of a Greenland glacier as it enters a fjord. To my mind the novelty in the manuscript centres around the near-ice helicopter CTD profiles and the balance of water mass and isotopic perspectives.

The topic of drivers and response of glacier melt is clearly important – in fact perhaps existentially so for coastal communities around the globe. This manuscript then focuses on a central challenge – a global impact has critical details at the ice-ocean boundary layer scale – just a few metres.

There is a claim of uniqueness regarding proximity to the glacier front (line 30). Possibly this is true, but there are a reasonable number of Antarctic studies with oceanographic work very close by floating icewalls (Fer et al. 2012; Stevens et al. 2014 among others).

The Introduction closes with a point that is pretty self-evident (lines 35-36).  “This study highlights the importance of considering the complex vertical and horizontal movements and transformations of glacial meltwater in understanding the freshwater dynamics in glacial fjords.”   Instead, I wonder if some more specific questions could be posed – perhaps around the horizontal extent of the boundary-layer zone, or something a more explicit in terms of the thermohaline budget?

Building on this – I would give the manuscript a new title to emphasize the aspects of the analysis that the paper does focus on rather than aspects it explicitly does not – i.e. the frazil formation. The manuscript actually says it could not be observed and will be examined in another manuscript – so why have it as the key point in the title? A very strange choice.  Furthermore, after the title “frazil” appears in the abstract and then not mentioned until section 3.4. Then the frazil section anticipates the heat budget section making for a very non-linear structure for the manuscript.

There’s plenty else here to highlight.  “Meltwater circulation and thermohaline budgets in a Greenland fjord very close to the terminal face of a glacier” would be my starting suggestion.

It also seems a lost opportunity to not examine more closely the connection between thermohaline and isotopic perspectives of water masses. No T-S diagram with isotope colours?  And then how does the observed partitioning compare with other estimates from this or other systems?

Does the paper address relevant scientific questions within the scope of TC?  The processes are all oceanic but driven by the ice-ocean interaction boundary condition.  The title suggests there is an element of frazil formation which would make it clearly in-scope for TC however as noted below the actual content doesn’t reflect this so distinctly. I would say it is borderline but certainly of interest to many TC readers.

Does the paper present novel concepts, ideas, tools, or data?  As noted above – there is clear novelty in the near-ice helicopter CTD profiles and the balance of water mass and isotopic perspectives.  Given the challenges of collecting a helicopter CTD profile I would have built on this to explore questions around what this brings us – does it sample within the boundary-layer structure?  i.e. where are the profiles in relation to the generally nice schematic of Fig 7?  Has all that effort brought some new insight?  The zoom in of the CTD data (Fig 2) seems to suggest those inner profiles are indeed different.

Are substantial conclusions reached?  Somewhat -  the authors provide clear estimates of the transfers in the heat and energy budgets.

Are the scientific methods and assumptions valid and clearly outlined?  Yes, mostly. 

• Were the two CTD units cross-compared?
• The description of the mooring data is very superficial – and are the data even used?
• Why use S_p and not S_A?

Are the results sufficient to support the interpretations and conclusions?   For the heat content material yes but clearly not in the case of the mention of “frazil” in the title.

Is the description of experiments and calculations sufficiently complete and precise to allow their reproduction by fellow scientists (traceability of results)? Mostly

Do the authors give proper credit to related work and clearly indicate their own new/original contribution?    Yes – maybe some additional references especially to Antarctic comparisons and an earlier reference for mixing line analysis.

Does the title clearly reflect the contents of the paper?    No – it is not clear why the frazil aspect is included in the title as it is misleading.

Is the overall presentation well structured and clear?

Somewhat:

• Fig 1 – panel b fonts too small, where were the profiles? Panel a fonts too small, bathymetry indistinct – possible to mask the land? Possible to expand b so we can see the full width of fjord?
• The profiles in Fig 2 are challenging to plot due to the wide range near the surface but the importance of small differences elsewhere… I suggest having a single panel that has the full T and S profile but just show the average or far field. Then I would have an additional figure with the zoomed in sections which, along with the isotope data, are the primary novel results – so make them big!
• Fig 3 – can the panels have an a and b? And the right hand panel seems hardly worth the zoom… instead I’d make the inset the 2^nd
• Fig 4 – because of the dominance of salinity are the density panels required? And the scales are a mess. I would have complete panels for T & S and then an additional figure doing the same for the upper 20 m – and this separate Figure could have different colour scale. As it is there are different depth and ranges for each. And for all that not much actually happens in most of the panels.
• Actually for Fig 4 – probably the other dimension worth highlighting in multiscale form is the horizontal – the helicopter profiles have great novelty so why not have some structure for the inner 100m?
• Fig 7 – why two panels? It would seem straightforward to merge. “Liquidus” is not mentioned anywhere else and presumably relates to the unfounded frazil conjecture. Why would the latent heat radiate as suggested in (a)?

Is the language fluent and precise?    Sufficient for clarity. A few minor suggestions:

• Lines 123-125 “distinct” and “distinguishable” needed in the same sentence?
• There are a couple of single sentence paragraphs that really should be either expanded or combined or removed. E.g. line 138 – I don’t think it is good practice to have a sentence that serves to identify a figure.. .how about… “ The temperature-salinity (θ-S) structure (Fig 3) reveals the key water types whereby XXXX”. (although see my point about Fig 3 made above).
• Line 140 – where is the “clustering” shown? Is it Fig 3?  Isn’t this just a scatter plot?
• Line 208 – another 1 sentence paragraph.
• Are fractioning and partitioning different?

References

Fer, I., Makinson, K. and Nicholls, K.W., 2012. Observations of thermohaline convection adjacent to Brunt Ice Shelf. Journal of physical oceanography, 42(3), pp.502-508.

Stevens, C.L., McPhee, M.G., Forrest, A.L., Leonard, G.H., Stanton, T. and Haskell, T.G., 2014. The influence of an Antarctic glacier tongue on near‐field ocean circulation and mixing. Journal of Geophysical Research: Oceans, 119(4), pp.2344-2362.

Comments on: Evidence suggesting frazil ice crystal formation at the front of Hisinger Glacier in Dickson Fjord, Northeast Greenland by Fleur Juliëtte Rooijakkers, Ebbe Poulsen, Eugenio Ruiz-Castillo, and Søren Rysgaard

by Sergei Kirillov, Centre for Earth Observation Sciences, University of Manitoba

In this paper, the authors examine an interesting dataset obtained in the marine-terminated glacial fjord in East Greenland and try to characterize the processes occurring at the front of the glacial terminus based on the obtained vertical profiles of temperature and salinity, and also some basic isotopes. 

Although I always support the idea of publishing the results of any campaign (especially conducted in the remote hardly or rarely accessible regions), I had mixed feeling while reading the paper. On one hand the presented dataset is worth publishing as it gives an interesting overview of the hydrographic conditions in the vicinity of the glacier terminus that may provide a new insight into the glacier-ocean interactions in the Greenlandic waters. But on the other hand, I was not convinced that the authors presented the confident lines of evidence to support their hypothesis of frazil ice formation near the terminus. My biggest concern is about using a very uncertain heat budget as a main tool to explain the positive anomalies of water temperatures observed within Polar Water layer at the distance up to 40 km from the terminus. The budget that does not include the speeds of inflowing and outflowing waters at different depths.

From my point of view, the observed anomalies have nothing to do with the frazil ice at all. My logic is: if the formation of frazil ice results in releasing latent heat that increases the water temperatures up to 0.5C within ~40-100m layer up to 40km away from the glacier, the melting of this frazil ice near the surface must lead to the opposite effect and results in consuming some (atmospheric) heat from the surface 0-10m layer. As a result, the surface layer temperature within that 40km range should be roughly 3C (=0.5C * (100m-40m) / 10m) lower than beyond 40km limit. However, we see nothing like this, but even opposite – the heat content of the surface layer within 0-40km segment is larger than further down in the fjord.

In Loewe (1961) paper (On melting of fresh-water ice in sea-water), the fast dissolving of the large chunks of fresh ice in salt water with negative temperatures (but 0.15C above freezing point at given salinities) was demonstrated. Even if one suggests that plenty of frazil ice crystals form at the boundary where fresh subglacial discharge meet cold Polar Water, what is a chance they survive long out of the subglacial plume next to the terminus wall? While the crystals stay within the plume with zero salinity and zero temperature, they don’t melt or dissolve. But once they drift to the salt water with temperatures above freezing, these small crystals will dissolve instantly (not as shown in Fig.7). Therefore, all these frazil ice processes (including a potential latent heat release) are likely occurring within a very limited distance near the glacial terminus and, from my point of view, can’t affect the water column at tens of kilometers away from the glacier.   

The second big concern is related to the suggestion that there are 3 oppositely directed flows within the upper 10m layer as presented in Fig.7. Please see my comment to Line 194 for more details.  

There is also a small general concern about attributing Dickson Fjord to the northeastern part of Greenland. I don’t know where the center mass of Greenland is, but its latitudinal span from 59N to 83N gives the center at 71N. From this perspective, the position of the fjord at 72N is at the east of Greenland, not northeast. 

Below is the list of other (mainly minor) concerns that need to be addressed:

Line 16:  A significant portion of GIS mass loss (88% (Mortensen et al. (2020))) can be attributed to melting at the front of marine-terminating glaciers (Straneo et al. (2012))

It’s not clear why two references are separated and what each reference is attributed to. If Mortensen et al.  found exact number, why do you need to include Straneo et al.? 

Line 20: Along with the increase in surface melt, the interaction of deeper warmer Atlantic Water masses with the glacier terminus has been found to increase submarine melting (Straneo et al. (2012); Zhao (2022); Holland et al. (2008)).

Submarine melting is also a “surface” melt. Just on the vertical surface. Wouldn’t it be better to specify that surface melt means the melt at the top of the glacier or use supraglacial discharge term?

Line 30: Change “glacier front” to “glacier terminus”.

Line 36: These findings contribute to ongoing efforts to accurately model and predict the impacts of glacial meltwater on the coastal region, ecosystem dynamics, and global sea-level rise.

The paper is focused on the frazil ice idea. I don’t think it is somehow contribute to the global sea-level rise problem.

Lines 43-44: Repeating information that was already presented in the Introduction.

Line 65: An oceanographic mooring device

    “Device” is needless here.

Line 75: Hydrographic sections of temperature, salinity, density and stable water isotopes were generated with a resolution of 1000 cells in both distance and depth directions.

    Mentioning the method of interpolation would be more essential than giving the grid size.

Line 121: Profiles, not transects! The insets in Fig.2 could be larger for better showing the observed anomalies (especially for salinities), I think. The size of empty space in the main panels allows this. There is an impression that the freshwater freezing line in the left panel of Fig.2 is above 0 at the surface that can’t be right. 

Line 129: “depending on the distance and location with respect to the glacier terminus”. 

You may be more specific (for the Results) by saying that temperatures generally decreased off the terminus and the salinities increased. The sentence in

Line 131 (“Interestingly, the transects show an increase in temperature in the Polar Water layer closer to the terminus”) is needless in this case. 

Line 133: “subaqueous surface melt”

Melt of what? If a glacier terminus is 0C, it won’t melt is seawater is above in-situ freezing point of, for example, -1C. The glacial ice will dissolve, but it’s a rather different process. If you consider GLACIER (zero salinity) MELT, you have to change “when ambient seawater is above the pressure-salinity-dependent freezing point” as follows: “when ambient seawater is above the pressure-salinity-dependent freezing point”.

Here the “surface melt” is used in relation to the vertical surface of the terminus, whereas in Line 20 the authors used this term for the melt at the top of the glaciers.

And it’s better to replace “subaqueous surface” with “sub-surface”. 

Line 148: “colder values” 

Change to “colder water” or “lower temperatures”

Line 148: “The Flights 1-3 values from the drone-deployed CTD interestingly exhibit colder values at salinities below 18, with the temperature staying relatively constant as the salinity increases up to 26, compared to warmer freshwater becoming colder with salinity at the stations 1 and 2 relatively nearby.”

The sentence is difficult to follow.

Line 151: “A runoff slope was plotted in Fig. 3 to investigate runoff mixing. Thermal satellite data showed that surface waters are released with a temperature of approximately 4 ◦C”

The authors should go through all the text and carefully check all used terms and different combination of terms. For example, what does “runoff” attribute to here? Coastal runoff? Runoff from the surface of the glacier? Subglacial runoff? What does “surface waters are released” mean? Released from where? There are quite a few confusing places like this in the text. 

Lines 153-155: The plotted “runoff slope” is also a mixing line. Why do you attribute the last term to the subglacial discharge only? Also, in “a ’mixing line … represents the line connecting the deeper water body…” what the deeper water body is? From the left panel, it looks like as the deepest level at stations 1-2 or ~170m (Fig.2), is it? Please specify, that this is the depth that approximately corresponds to the draft of the tidewater glacier. If you do so, it will be easier to understand the following sentence: “Therefore, the presence of subglacial discharge in the deeper water body is unlikely.”

Line 156: “From the meltwater slope plotted in Fig.3”. 

There is no such line in Fig.3. There are “runoff slope”, “mixing line” and “surface freezing line”. Please, be more consistent with the terminology you use and choose it carefully. 

Line 158: “In Flights 1 and 2 data, however, the slope changes from positive to negative in the salinity range 24-28. This slope becoming positive can be attributed to meltwater influx, as these lines change direction towards the subglacial discharge point (θ, S) = (0 ◦C, 0). Interestingly, in Fig. 3, a slope change occurs in the lower salinity waters, with the lowest salinity water slope parallel to or even steeper than the runoff slope.”

This is very unclear. Slope of what? If it is what I understood, I see no slope change between 24-28. It is still negative at both <24 and >28. You probably meant a positive anomalies of temperatures within this range of salinities? And again, what does meltwater influx mean? From the surface? Subglacial? Coastal? What does “these lines change direction” mean? 

Line 163: “To increase spatial distribution over the observed differences”

I don’t think “increase” is a good word here. Maybe “to examine the spatial patterns of T, S, D in more details”? Figure 4 needs station marks showing the dates of different stations in different color. There had been up to 2 weeks interval between different CTD casts. I would also think of adding few isotherms between -1 and -1.5 (-1.3 an -1.1 for example) to the PW layer temperature panel. It will emphasize the gradual increase of temperatures toward the glacier.

Line 166: Why do you use “contour plot” here and “sections” for T/S/D earlier?

Line 175: “Interestingly, this deeper signal becomes (slightly? as compared to the depletion in the surface layer) more depleted further from the terminus”.

Line 183: “Below this decrease, a second depleted signal was found at around 10 m depth, which no longer would include glacial runoff to the surface ocean, but due to the positive ocean temperatures can still be caused by surface melt (Fried et al. (2018))”

I found it difficult to understand what is a role of ocean temperature in forming this depleted layer.

Line 187: “…there is still a depleted water signal distinguishable in the Polar Water layer. While this meltwater signal could be attributed to subglacial discharge, the salinity profile does not show a corresponding freshwater signal”.

Are you talking about Polar Water near the glacier terminus? Because if so and if the inset panels in Fig.2 were larger, it would be easier to see that PW water near the glacier is slightly fresher (the depleted dO18 signal is not very strong either!). 

Line 194-199:

I don’t find this explanation very convincing. First of all, I can hardly accept the idea that the difference of T/S/dO18 observed within upper 10 meters is related to the classic estuarine circulation in fjords. Of course, the current velocity data (from the mooring?) would help to examine this theory, but I would suggest a slightly different scenario. What if the entire 0-10m low-salinity surface layer is formed at the end of winter because of sea ice melt. Later in the summer season, it gets heated up to 9-10C and, at some point, starts to receive the dO18 depleted water with zero salinity discharging from the top of the glacier (or maybe from the land). This depleted water could explain the signal at 1m. What is happening below 10m is the PW that flows toward the glacier at 50-75 m (see Fig.4 temperature panel. Adding -1.3C -1.1C contours to that panel will help to better visualize this I believe). This water mixes with subglacial discharge and upwell towards the surface. However, at ~10m depth it meets the strong density interface and spreads off the terminus giving that depleted subsurface tongue at 10m. For me, this scenario looks more realistic than your idea of an opposite estuarine circulation (with 3 oppositely directed flows!) between 1 and 10 m depths. 

I would strongly recommend to show the SW part in Fig.6. If the scenario I suggested is correct, we may see some positive thermal signal at ~10-20 meters that spreads off the glacier. 

Line 273: “temperature and salinity” are not used in the formula (4)

Line 277: I think the radiation was measured not at the sub-surface mooring, but at the coast.

Line 287: “Subtracting these values from the incoming radiative heat flux”

Was air temperature below surface water temperature in August? Because if not, one needs to add sensible heat flux to the surface energy budget, not subtract.

Below are my answers to some specific questions the reviewers are supposed to address

Does the paper present novel concepts, ideas, tools, or data?

The paper accommodates a piece of novelty related to using of a new drone-driven CTD in the vicinity of glacier terminus. However, the majority of data used to build the story is related to the standard measuring oceanographic techniques. Although the basic idea of attributing the anomalies in water temperatures to the frazil ice formation is interesting (but not new according to the authors), the lines of evidence don’t look 100% convincing without more detailed and specific observations (i.e. accumulation of frazil ice in front of glacier in winter time, current velocity measurements for estimating Polar Water residence time in the fjord, or something else).

Are the scientific methods and assumptions valid and clearly outlined?  

Mostly yes, although there is some unclarity in terminology that has been used in the different parts of the paper. See my specific comments above. From my point of view, the most problematic part in the paper is 3.5. I honestly tried to get through it but failed. The authors should consider put some additional efforts for making this chapter clearer and easier to follow. A combining heat fluxes (within specific period of time) with the observed heat storage in the water (without knowing the history of that water and its residence time in the local system) is a very tricky game. 

Are the results sufficient to support the interpretations and conclusions?

I don’t think the presented results are sufficient to support the main hypothesis about the presence of frazil ice. At least in quantities that would explain the observed positive thermal anomalies within Polar Water layer up to 40km away from the glacier.  

Does the title clearly reflect the contents of the paper?

Yes, although the content of the paper didn’t convince me that the main result shown in the title is correct. 

Does the abstract provide a concise and complete summary?

Yes.

Is the overall presentation well structured and clear?

The structure of the paper is fine, but the authors need to carefully revisit the terminology related to the different portions of the glacial discharge and melt.