This study presents a multi-proxy record from a marine core collected from outer Petermann Fjord (PF), NW Greenland, to reconstruct Holocene sea-ice dynamics in this region. This reconstruction builds upon previously published geochronology and sedimentological analysis conducted on the same core (Reilly et al 2019) and provides new insight into regional sea-ice dynamics and the influence sea ice likely had on the presence/absence of an ice tongue fronting the Petermann Glacier (PG). By comparing this new record with existing records in the region, Holocene sea-ice reconstruction in Nares Strait has been produced. The results indicate the important role sea ice and ice arches located in the north and south of the Nares Strait had on landfast sea ice within the PF and therefore on the presence or absence of the stabilising ice tongue of the PG. The results and discussion add an important perspective on the role of the marine cryosphere in stabilising local ice tongues in NW Greenland, which in turn has implications for modelling future behaviour of marine-terminating glaciers and dynamics of the Greenland Ice Sheet.

Overall, the article is well written and presented. The results and discussion are robust and support the interpretation and conclusions made. It is evident that a significant amount of analytical work has been conducted and the results are carefully considered.  The conclusion that landfast sea ice is an important control on ice-tongue presence/absence, although not unexpected, is an important outcome and is within the scope of The Cryosphere. I do not have any major comments, and I would recommend publication after addressing some minor points listed below:

Specific comments (intermediate). Numbers link to the manuscript line number.

I do like the short title, but I think you could expand upon this to include the broader regional implication of the study with regards to Nares Strait sea-ice dynamics. I suggest you adapt this to indicate the wider reach of this study.

Figure 1a: can you add /label the deeper Atlantic water circulations on the map?

Section 2: Water depth and the location of sills appears to be an important control on the AW, but it is not clear where these sills are, in particular the 220 m deep sill in Kane Basin. Can you add either a bathymetric contour or sill on your Fig 1 b? or if that looks cluttered add a little bit more detail in the text.

115: can you add the current temperature of the modified AW and where it is located in the water column? 

117: in this sentence you state ‘This sill impedes the throughflow of the densest AW…’ do you mean it prevents AW from the south altogether or is there a density gradient in the AW? If the former I suggest you rephrase to ‘…impedes the throughflow of the dense AW…’

137-140: Can you add a bit more detail for Fig. 2 in this section, such as years where landfast sea ice formation in PF acts independently to Nares Strait, and the sea-ice conditions at the time of the ice-tongue calving events?

140: add reference and detail on how productivity changes with sea-ice conditions. Also, how does Supplementary Figure 1 show PF with ice edge conditions in spring/summer? Can you add an example year in the text in line 142 ‘…during years (e.g., ).

Section 3.1, pg 7: I would like to see a bit more detail for the core description added in this section. In line 179 and 180, can you add more detail on the ‘…coarser particles. ‘ and ‘...coarse material…’ respectively, such as rough grain size? Also, can you add where the core is laminated in this section rather than in the results?

Section 3.3 Sea-ice biomarker methodology is very detailed. Does all this information need to be in the main text? Can some of it be moved into the supplementary section e.g., Table 2?

Section 3.3: Foraminifera abundances are interesting and provide potential information on productivity, but assemblage data would also be very insightful with regards to the sea-ice or ice tongue presence in the PF as well as the presence of AW (e.g., presence of the foraminifera Cassidulina neoteritis). Is there a reason assemblage counts have not been included here?

366-375: this section describes the DIP₂₅ results, but this is in the Supp info. Is there a reason for this or can this data set be moved into the main text?

General results section comment. The core stratigraphy and environmental interpretations are sporadically included within the results section. In doing so there is some repetition e.g., line 332 and 349 and new information on the core is added intermittently (e.g., lines 353-354). I would suggest removing the interpretation and sedimentology detail out of the results section unless it is needed to justify why data has not been included (e.g., lines 329-331). Instead, I would like to see this detail added to the methods section (3.1) when the core is initially described and within the discussion section where the multi-proxy results can be linked to the core interpretation.

Lines 470-475: could another explanation for the lower terrestrial input during the re-establishment of the ice tongue be related to grounding-line proximity to the core location?

Figure 5h: the reconstructed ice-tongue extent (km) at ~590-600 cm doesn’t seem to match that produced in Figure 5 of Reilley et al (2019). I also suggest 5h extends back to 555 cm to match the datasets for this study.

518: The author state that increased meltwater runoff may have led to sea-ice instability. Can they add a bit more detail on the process and/or reference?

Section 5.3, line 593-595: an assumption made is that dependent on ice arch configuration PF and Kane Basin will have opposing sea-ice conditions. What is this assumption based upon? The 4 satellite images in Supplementary figure 1 or published observations? If the latter then can you add the reference to the text? If this is based on the authors' observations, it would be worth adding a figure like Fig 2 that includes Kane2b sea-ice conditions to show this modern relationship and that there is indeed opposing sea-ice conditions to strengthen this assumption. This could be added to the Supp info.

640: This is the first time that ‘The arrival of little auk colonies…’ have been mentioned. Can you add their significance with regards to understanding sea-ice conditions?

654-662: Enhanced seasonality is an interesting outcome. Are there any other records (marine/terrestrial) that also indicates enhance seasonality to support this interpretation? Why would seasonality be enhanced during this period? You also start the sentence ‘Possible reasons…’ what other potential reasons are there for this and why is seasonality your preference?

661: What historical records are you referring to?

731-732: Here the authors proposed the formation of seasonal ice melange in the fjord shortened the calving season. What is the evidence of this? Can you refer to the appropriate results?

741-743-: I found the sentence starting ‘Thus, this scenario…’ hard to follow. Can you rephrase this?

Specific comments (technical).

41: add a, after Historically.

Figure 1A: there are purple currents that have not been labelled. Can you state what these are in the caption?

108: remove ‘the’ before Baffin Bay

109: capital C for Robeson Channel

134: remove ‘vast’

195-197: add ref to end of the sentence

Section 3.3 Planktonic and benthic foraminiferal abundances: change to 3.4

298 and 300: add ref to end of sentences.

Line 391: change to ‘These become especially apparent in the dinosterol and campesterol concentrations between 400-500 cm and the β-sitosterol concentrations (add core depth)

447: Add Washington and Hall Land onto Figure 1c

544: I suggest you change the wording ‘…bear witness of…’ to suggest prolonged

560: I suggest you do not start the sentence ‘Especially…’. Also, it is not clear from this sentence whether both declines in IP₂₅ fluxes are accompanied by a decrease in TOC but the second is more significant, or only the second IP₂₅ decline is accompanied by a decrease in TOC.

602 and 603: be consistent with landfast or fast.

613- 615: I suggest you rephrase this sentence to ‘Thus, abundant driftwood delivery to Ellesmere Island/northeastern Greenland together with the formation of beach ridges is indicative of seasonally open waters along the coast (Fig. 6A).

Figure 8: A simplified version of Fig. 6 is not needed. I suggest you remove Fig. 8 and add the ages to each panel in Fig. 6

Supplementary Figure 1: can you include what ice arch state each image is showing in the caption and label each satellite image a), b), c) and d)?

Figures: overall the figures are nicely presented. My only minor criticism is that the y-axis does not line up to the graphs. Whilst I appreciate this is done to enable the graphs to fit on one page, it does make it harder to see the values and which axis relates to which graph. In some cases, the lower axis overlaps with the overlying graph and the overlying graph is covering some of the data points e.g., Fig 4e. Can the authors find a way to make this easier for the reader to view?

This study aims at understanding the role of sea ice on the stability of the Petermann Glacier floating ice tongue. It is based on a series of sedimentary and biogenic proxies preserved in a marine sediment core spanning the Holocene and collected 80km from the 2015 glacier grounding zone. It focuses on sea ice-glacier-ocean interactions at one of the key regions of freshwater export from the Arctic Ocean. The purpose of the work is clearly articulated, and the multi-proxy approach used is suitable. The authors make a thorough review of the oceanographic setting and signal associated with each individual proxy used, ensuring that the paper can be understood by a broad multidisciplinary readership.

The context and results of the study are presented in a clear and logical manner. However, a key component of the interpretation may need to be revisited. The authors compare their marine sediment core to other regional proxy records, including core AMD14-Kane2B (Georgiadis et al. 2020), located in the Kane Basin to infer the role of the northern and southern ice archs on the sea ice conditions in the Nares Strait. The interpretation of conditions in the NOW and southern ice arch are at odds with what is reported in the paper cited (see also comments below, where specific sections have been identified). In particular, the interpretation of the intervals of stability/instability of the southern ice arch is not supported by the available data. As a consequence, the interpretation that the inception of Petermann ice tongue is preceded by the transition towards a southern ice arch regime is not corroborated (for the period 3-2.2 kyrs BP) by the only data available for the southern ice arch region. This aspect of the study should therefore be toned down. For this reason, the paper requires moderate revisions (note that the system only allows me to select minor or major revisions).

General comments:

• The paper is lengthy and unnecessary repetitions between sections should be avoided (e.g., Introduction and Regional settings).
• Neoglacial should be capitalized

Figure 1.

• Please precise what the extent of the NOW represents (average between year-a/year-b, and season). It is not necessary to repeat the same sentence twice “The approximate extent of the NOW is indicated with a black dashed line.” The red box already indicates the close up.
• From the caption, one may interpret that core AMD15-Kane2b was analysed as part of this study. Should add reference to paper.

Line 107. Since this sentence is already the first of the introduction, it should be left out here. Avoiding repetition will help reduce the length of the paper.

Lines 150-152. This sentence does not read easily. Please try to reformulate.

Line 205. Should also mention HBI II-producing species. Co-production implies same species, strictly? Please clarify.

 Figure 2.

• Line 166 should be Hall
• “After 2012/2012, the differences increase due (…)” remove “s” in “increase”.

Lines 222-225. This information does not seem necessary since both sterols are not direct indicators of sea ice conditions. Could only keep the last sentence of this paragraph.

Lines 234-236. These sentences seem out of place. This information (storage of research material and subsampling) should be included in the first paragraph of the Method section.

Line 252. “The d¹³C_org reproducibility of non-acid pretreated (…)” add “of”

Line 284. “Dry bulk densities were calculated (…)” this is repetitive with Line 258. No need to repeat multiple times. If applies to all, could be included in the first paragraph of the method section.

Line 288. Could rephrase to “A multitude of environmental factors determine the abundance and species composition of benthic and planktonic foraminifer assemblages”

 Line 294. Replace light conditions by light availability.

Results section

• Verify significant digits for the foraminifer fluxes.
• Why aren’t the results reported against the age rather than depth (or both)? This would make it easier to follow the changes/story in the context of Holocene climate variability.
• The same structure (i.e., order) could be maintained between the Method and Result sections.

Line 317. Add “s” to “unit”; same in Line 342 (please correct throughout)

Figure 3.

• Should include an age axis.
• Remove “s” in foraminifers results
• turquoise
• Should mention in the captions why the fluxes do not extend beyond 400 cm.
• Should include HBI concentrations normalized to TOC.

Line 441. “Sedimentary unit 3 represents” add a “s”

Lines 447-449. This sentence is not clear.

Figure 5. The title of this figure is “Temporal changes in the environmental conditions in Petermann Fjord”. Yet, there is no indication of “environmental conditions” in this figure, with the only exception of the ice tongue length from the grounding zone. To make this figure truly distinct from the other figures presented in the results, more interpretive information could be added.

Line 600. “In years when only the northern (…)” replace were by when.

What about pelagic productivity?

Figure 6. Great representation of scenarios.

Lines 695-704. This section as well as the interpretation of the southern/northern ice arch conditions (e.g., Line 738), need to be revisited. The IP₂₅ fluxes reported by Georgiadis et al. 2020 are very low around 2,200 cal yrs BP. Authors from this paper indicate that IP₂₅ fluxes are at their lowest between 2.2 and 1.1 cal ka BP, which they interpret to result from an unstable southern ice arch in Kane Basin from 3.0 cal years BP. Little auks are present in low numbers at lake Annikitsoq and only one discrete event of local peak in bird abundance is reported from lake Qeqertaq (Davidson et al. 2018). These data do not point towards a stable spring/summer North Water polynya and return to a southern ice arch dominated regime in Nares Strait, as suggested here.

Lines 760-762. See comment above. Perhaps supported for the rapid extension at 600 cal yrs BP and double-arching, but the evidence from Kane Basin is pointing towards less stable southern ice arch from 3.0 to 1.1 cal yrs BP.

This study brings together a large amount of data from a range of proxies to investigate the dynamics of the Petermann Glacier through the Holocene – in particular the dynamics of the ice tongue. It then sets this in the wider context of other studies in the region. The paper is generally very well written and structured, though I found the discussion sections rather long with quite a bit of repetition. This may be due to the authors trying to bring together such a wide range of data. I think it might be helpful, perhaps, to concentrate on the key aspects of interest.  

There is a very clear and detailed introduction providing background context for the study. I think it would be useful to include on fig 1 other cores in Petermann Fjord that have been used in the Reilly et al (2019) paper to assess the development of the ice tongue. There is considerable reference to the findings from this earlier paper, so it would be helpful to see exactly where the additional cores are from. This is particularly relevant as this earlier paper is used to help support the timing of ice tongue development.

Methods provide a detailed rationale for the proxies used – very clearly outlined and justified. The datasets are then very clearly described in the results section. There are numerous well developed figures used to illustrate the data. One aspect that it would be useful to consider adding is a plot showing the development of the age model – even though this may have been published elsewhere, I think it would be useful context here.

The data are plotted against depth in many of the figures (3, 4 and 5) – this is fine for the initial results description, but I think it would be clearer if the data were then plotted against age to help in the discussion section. Fig 5 shows an age axis, but it is not always easy to follow.  

Discussion

I found Section 5.1 confusing in places - it seems to assume an interpretation that is more clearly presented in section 5.2 – and often based on the earlier Reilly et al. 2019 paper. For example in lines 463 – 464: ‘Following the break-up of the deglacial Petermann ice tongue at ~6,900 cal yrs BP (unit 2/unit 1C boundary)’ ok, this does seem sensible, but would be good to be clear what the evidence is for the break-up of the ice tongue based on the data presented here (based on IRD input, disappearance of laminations, increase in marine productivity?).  

Development of the ice tongue (Lines 468 – 490) - It is not clear what the evidence is for the development of the ice tongue again from 2,100 cal BP (transition from unit 1C to 1B). Again, this is partly based on Reilly et al 2019 and supported by data presented in this paper, but this discussion comes in a later section. The decrease in marine productivity from 600 cal BP seems clear support for the development of the ice tongue…

I wonder if section 5.1 is actually needed as a separate section? I think it would actually be clearer if this section was removed and key parts subsumed with the rest of the discussion section. This would likely save on repetition and make the paper easier to follow.

Section 5.2 presents the interpretation of the dataset very clearly and would actually be better coming before section 5.1 (or as suggested above remove section 5.1). Perhaps be clearer what the evidence is for initiation of ice tongue development from 2100 cal BP – based on Reilly et al and other cores closer to the grounding line (ie up-fjord)?

The discussion sections are very detailed, and from what I can tell they seem to be well supported by the datasets presented here. I think there is probably scope for reducing the length of the discussion (avoiding repetition). I think having the key datasets presented together plotted against age rather than depth would also help the reader. Fig 7 presents some of the data form the paper plotted against age, alongside a range of other datasets – this is a really useful to visualise the various datasets, but having more of the primary data presented in this paper plotted against the age model would be helpful.

This is clearly a very important contribution to our understanding of the dynamics of Petermann Glacier, and summarising an impressive amount of data. It is clearly worthy of publications after some minor amendments.