REPLY TO REVIEWER 1: BRINER

Review of Cavnar et al., “In situ Cosmogenic 10Be and 26Al in Deglacial Sediment….”

Responses to reviewer comments are italicized below.

Cavnar et al produced 10Be and 26Al measurements in sediment deposits sampled from varying distances along paleo-flowlines of the former Quebec-Labrador dome of the Laurentide Ice Sheet.  Varying amounts of nuclide inheritance in both modern sediments and early Holocene landforms allow the authors to make inferences about limited glacial erosion in this region during past glaciations. Based on prior work in this and other ice-sheet dome centers, it is expected to find inheritance signals; however, the exact quantities may be relevant for crn tracer studies in offshore sediment cores.

That this paper does not relate to geochronology leads me to be somewhat neutral about the journal choice.

I liked this paper a lot. I enjoyed reading it and found it grammatically clean. The results were interesting to read about and mull over. The figure and table support is solid.

I have several comments that the authors could consider.

We appreciate your thorough review of our manuscript. 

1) This is great work, I wonder if the pitch for relevance to offshore work could be amped up. The title could be more informative, for example, by including a conclusion. As I write in my synopsis above, it strikes me that the findings are not highly novel, perhaps even only mildly informative, when couched in terms of inferences on past ice sheet exposure/burial/erosion. I don’t think anyone would expect a story from the sediments to be any different than that from surface boulder and bedrock crn data, in addition to other geomorphological datasets, that already exist. That said, it seems to me that the real new information, and more important motivation for this work, is to inform new offshore sediment records like those of LeBlanc et al (2023) and others. To this end, I would suggest being more up front about this motivation (like including this impetus in the abstract and giving it more emphasis throughout the paper; it’s there, but I found it not front and center).

This is a good point and we appreciate your perspective. We will work to further emphasize the significance of using dual-isotope cosmogenic nuclide analysis to inform offshore sediment records in the next draft of the manuscript. 

2) I don’t see how the Al/Be data in this study can be used to support claims that ice persisted throughout interglacials (and I sort of disagree that the community thinks that such persistence would be unexpected). It seems to be a conclusion from the LeBlanc work, but the data in this study standing alone, I don’t see the evidence. In fact, when considering the Al/Be ratios at the proximal deglacial sites (5.2±0.8) against Figure 3 in LeBlanc et al (2023), it seems more plausible for every interglacial to have experienced many thousands of years of exposure. If making this point and further supporting LeBlanc is important to the team, I suggest some exposure/burial history modeling to see if these crn data uniquely support persistent interglacial ice cover. If not, it just seems like rather unsupported bolstering of a previous finding by this research group. Either way, I think the study stands just fine without pushing this conclusion.

We agree and will reconsider the paper’s focus in revision, hewing more tightly to the sediment history theme.

Line by line comments

13, don’t really need the word ‘buried’ here

That’s a fair point; we will remove buried. 

18, ‘surface erosion’ of bedrock? Or of sediments, too?

Surface erosion of sediment that we assumed is carried by modern river systems. We will be more specific here and elsewhere in the ms.

22, ‘average several thousand years’.  The average and standard deviation of other measurements are reported numerically; I suggest doing the same with this one – that is, report the average and standard deviation of inheritance in terms of years.

Noted, thank you for the suggestion.  Will make this change.

22, could remove words from ‘after correction..’  I find this multiple paragraph abstract long and wordy. I think it could be made much more succinct. Readers can find the details in the manuscript body.

Noted on the long abstract. We will condense it. Correction is an important term because it indicates the raw data have been transformed. We will make sure it’s in context. 

31, could just say ‘reported by others.’ ‘reported here’ is a little redundant in this sentence since the first part states that there is inheritance in this study’s dataset.

This would definitely improve the flow of the sentence, thank you.

44, I don’t think anyone honestly thinks land-based ice domes like these ‘collapsed’, suggest other wording. Also, find other wording for the ‘northern Canadian ice domes’. What are these, like the Barnes and Agassiz ice caps?

We will reword this for clarity. 

48, ‘Because’ sentence.. suggest rewording. ‘far north’ but then study Rice study, which is arguably the southernmost of the three LIS domes. Be specific.

We will fix this, thank you.

58, to one of my main points, this paragraph closes to say results could be important for offshore work. I think this paper would be stronger this point was more directly as a major impetus of the work. I guess after reading the entire manuscript, I felt that the work was done as a companion for offshore studies, but I did not get that sense while reading it.

Thank you for pointing out that the connection to offshore work is not clearly stated in the paper. It was a strong motivator for this study so we will make sure to edit the manuscript accordingly. 

80, section 2.1, this is subjective not objective comment, but covering the basics of Al/Be methodology seems potentially below the readership. Critical to include in student writings, but not necessary in disciplinary manuscript.

Thank you, we will consider this. 

115, reword, not sure ‘for inception to deglaciation, the LIS was…’ makes grammatical sense.

We will reword this, thank you. 

122, 137, the content in these two paragraphs are the same, but are separated by a paragraph on a different topic, suggest combining and streamlining section 2.2

We will reorganize section 2.2. 

126, suggest write ‘not as well constrained by comparison’

This is a good addition. 

128, I’m not sure I would characterize this as “it is debated.” The Zhou and McManus reference is missing; I didn’t look up what they wrote.

We will revisit the wording. Thank you for catching the missing reference. We will add it. 

131, What is meant by “these”? Prior sentence cites LeBlanc but the data aren’t mentioned in order to justify use of “these”

We will reword for clarity, thank you. 

135, ‘commonly held assumption’ I disagree with.  The Holocene technically isn’t fully ice free (eg, Barnes ice cap contains Pleistocene ice) anyway. As a rule, I don’t think it’s entirely responsible to push a controversy to elevate importance of one’s work (I wince a little reading this after I wrote it, but I do think it can be said here, sorry if harsh). And as LeBlanc points out, one look at those mild middle Pleistocene interglacials, why would one think that every interglacial was absolutely ice free in the first place? I, for one, think the persistence of small ice caps/domes in the three LIS centers, like today on Baffin but even a larger Barnes, during may Pleistocene interglacials is actually highly likely.

Thank you for pointing out the potentially inappropriate and inaccurate nature of stating something as a ‘commonly held assumption’. We will revise this section to better convey the lack of certainty surrounding Pleistocene interglacial ice cover opposed to suggesting to the reader that being ice free during these periods is a common assumption. 

166, who’s ‘they’?

Will change to ‘this study’. Thank you. 

197, after reading this section, some relevant research could be included. I would think that the abundant literature on bedrock crn inventories from ice sheet settings showing inheritance is worth citing. As are locations within ice sheet boundaries where pre-Holocene sediment has been preserved (Miller et al last interglacial sediments in lakes on Baffin and Greenland, and maybe look into old sediments preserved in the Manicouagan Reservoir, I seem to recall a presentation showing old seds right there in the heart of Labrador).

Great suggestions. Thank you. We will reach out if we have questions about these sources. 

234, love this sentence, to repeat the above, would add emphasis to this earlier in the paper

Noted. See reply to first comment. 

Figure 1, Panel A says “text” in N Atlantic. Box showing area of B is incorrect. Overall could be improved significantly, make larger for starters, no need to be so small – really important figure here.  The dashed ice sheet extent lines are pretty difficult to decipher (why dashed lines?). Panel B is crucial for this work, should be improved, put on shaded DEM perhaps, maybe even color, would be nice to see elevation of the transects and the overall terrain. Dome center does not coincide with ‘last ice location’ - why not? Wouldn’t the retreat center (location of last ice) make more sense if interested in exposure/burial durations?

Thank you for catching this typo; we will get rid of ‘text’. We will also create a more spatially accurate inset for panel B. We will definitely increase the font size and can add a DEM to panel B. Dashed lines can be changed to make them more decipherable. 

The dome center does not correspond with ‘last ice location’ because the isochrons date beyond the deglaciation of the Laurentide ice sheet—to where ice is only present in Greenland and parts of Baffin. We only added the isochrons visible in our panel B study extent to the legend. But, we can see how this is confusing to the reader and can add all the isochrons visible to a new legend in panel A. 

322, would be nice to see in this table - the % correction for Holocene exposure. So many samples are from generally great depths, I would not expect there to be much correction, still, would be nice to show to readers.

We will add this in, thanks. 

361, ‘with distance from center’ note comment above, I question the choice of ‘center.’ Furthermore, distance from center is a little awkward, it only makes partial sense. Plotting distance along flowline would be better.  Or by retreat age would be best.  Figure 3c shows two deglacial samples within the 9 cal ka ice limit that have ratios >8.  In the south transect, samples in similar positions (just within 9 ka limit) have ratios of 5.5 and 3.5.  Seems like data plotted along paleo-flowline distances would make most sense?

We chose to plot samples according to their distance from the estimated center of the ice dome (estimated based on sources such as Couette et al., 2023; Dalton et al., 2020) because of the general radial retreat from the coastal margin to the dome center. We wanted to spatially analyze 26Al/10Be to see if it fit with the overall assumption of the center of the dome being the most persistent ice/last to deglaciate. However, your comment brought us to re-examine the shape of the isochrons in Figure 3. We can also plot the distance from the estimated 9 cal ka isochron. 

Figure 5, could be two side by side plots, one with modern and one with deglacial. Would be cleaner, for example, to symbols don’t look different in grayscale (how I printed it out). Also suggest for all plots, sticking with blue/red/purple color scheme in map figures, why not carry that color scheme over to the data plots?  Can you add the bedrock sample?

Great idea to do side by side plots for modern and deglacial samples and to keep the color scheme consistent throughout all the figures. Thank you for pointing out the poor symbol distinction in grayscale. 

403, I find it awkward to take an ‘all sample average’ for samples that are at different distances and with obviously different exposure/burial histories. It only occurred to me later that you would even bother taking a total average because you could consider this an integrated value for what might make it to the ocean. This is one example of how stronger emphasis on the offshore work (and included as a justification in the abstract, if not the title) might make more of your thinking here in the thick of data reporting make better sense. To me anyway. In the abstract, upon first reading, I was like, ‘why in the world would you average this value?’

We will definitely add the justification for taking a total average and emphasize its relevance as a value comparable to IRD 26Al and 10Be concentrations. 

404, neither 6.1±1.2 nor 6.6±0.5 are statistically different than 7.3±0.3.

That’s true, they are lower but are not statistically significantly so. We will add this to the end of the sentence and report the p value so it is clear to the reader. 

405, ‘consistent with burial after initial exposure’ It could also be that the sediments themselves were emplaced at time zero with a burial signal. Isn’t that the nature of sediment work, you don’t know what’s post deposition vs. pre deposition? Or?

We can rewrite this sentence to make it less confusing. Yes, the sediments themselves were likely deposited with an inherited signal from a previous interglacial exposure. However, we characterize deglacial sediment samples as having a ‘depressed’ ratio and modern sediments as having ratios at or near production value because of modern surface exposure. We took an error-weighted mean for both sample types (Holocene-corrected deglacial and modern-fluvial). We expected the mean 26Al/10Be for deglacial sediments to be ‘depressed’. However, the modern fluvial sediments also had a ‘depressed’ error weighted mean. This sentence was supposed to be a lead-in for our further discussion of sediment sourcing in section 6.3. We will reorder sentences in this paragraph to match the order of sub-heading topics in the discussion section.

Figure 6, love seeing the LeBlanc panel here, again emphasizes the importance of this work for the offshore stuff. Panel C  I don’t see any justification for the splitting of these samples into two groups. Why not three groups, or five groups? What is the threshold that allows distinction of “proximal” v “distal,” and again consider distance as was measured versus flowline distance.

Thank you. It helps us to know that from an audience perspective, the LeBlanc panel fits well with the rest of the figure. We decided to split our deglacial samples into two groups because of the visual split in 26Al/10Be ratios plotted against distance from the center of the ice dome. It seemed that between 300-350 km from Labrador City, the ratios jumped to being much higher, opposed to a gradual uptick with increasing distance (Figure 5). However, we do see how it could be confusing to the reader to split one sample type into two lines on this figure and will consider other ways of expressing this spatial difference in ratios.

465, ‘cosmogenic dating’ seems lazy phrasing

We will replace this with a more descriptive heading. 

469, Briner and Swanson 1998 was neither on the LIS or Fennoscandia/Greenland.

Thank you for catching this. Changed to  ‘landscapes in Fennoscandia, Greenland, and near the margin of the Cordilleran Ice Sheet’. 

472, There are probably >1000 of crn data published from Baffin, some papers included much data on sediments (boulders, cobbles, pebbles; some of mine, reach out if you have questions), why point to one cobble result from the Davis study?  Throughout this paper, there is a bias toward citing authors’ own work. To some degree this is difficult to avoid, certainly I’m guilty too at times.

We will review more literature on crn data from Baffin Island and add sources to this paragraph. We will also ensure that a wider range of studies are cited.

483, this paragraph and previous paragraph, difficult to summarize all this work in 1-2 paragraphs. That said, the tactic taken, to cite individual studies and individual samples, seems destined to be highly exclusive. Is there a way to discuss this more broadly and lean more on synthesis papers?  It’s kinda like the main author knew a few papers really well, so leaned toward citing those again and again instead of a more comprehensive view of the literature. Of course this is easy for me to write, and much more difficult to implement, but I’ll throw it out there.

We will make sure to reference more synthesis papers to bolster our review of the literature. We heavily cited Ullman et al. and Couette et al. throughout the paper because of their relevance to our specific field area and comparable methodology. That being said, we can definitely incorporate more references to synthesis papers in lines 483-493. This would fit well considering the paragraph examines cosmogenic nuclide inheritance throughout the northern hemisphere. 

507, not really ‘remnants’ per se. If ice shrinks to become smaller but in balance with a new climate state, does it make it a ‘remnant’?

Interesting point, we will change the description to something more appropriate. Perhaps ‘retracted ice sheets must have lingered…’

507, ‘must’ seems strong. Language in the LeBlanc paper uses no such strong language, careful to rewrite history.

Thank you for catching this, will change to ‘possibly lingered’. 

515, replace ‘taken from outcrops’ with ‘bedrock’ (could add “eg” in front of Marsella ref, again back to comment about citing authors’ own work a little heavily).

Thank you, changed to ‘bedrock’. 

526, I do not believe your data allow you to make these conclusions. See earlier comment at top of review. Concentrations and ratios of ~5 can be easily explained with burial during glacials, even figure 3 in LeBlanc shows this. If there is anything close to a flaw in this great paper, it is here in my opinion. Unless you perform some modeling that says the data can and are uniquely explained by ice cover persisting throughout during interglacials could you use language used in this paragraph.

Good point that we have considered and in revision we will present a range of ways in which low ratio sediment can be generated including ice cover but not mandating it. Alternative scenarios include sediment storage on the landscape in meters thick till sheets and even thicker deltaic deposits.

561, to repeat, I do not see this statement being supported by data, and I fear this work will be cited in the future to support such a notion, but careful reading reveals no such support.

Thank you for making us aware of this. See earlier comment. 

REPLY TO REVIEWER 2

Review of Cavnar et al., “In situ Cosmogenic 10Be and 26Al in Deglacial Sediment….”

Responses to reviewer comments are italicized below.

This manuscript by Cavnar et al. provides an interesting and novel framework for applying cosmogenic nuclide exposure vs. burial dating to understand the behavior of the former Quebec-Labrador Ice Dome of the Laurentide Ice Sheet in eastern Canada. The authors measured in situ cosmogenic 10Be and 26Al concentrations in both modern river sediments and deglacial landforms across a ~1000 km transect and found 26Al/10Be ratios indicative of limited erosion and nuclide inheritance, which led them to suggest that the Quebec-Labrador Ice Dome could have persisted through some interglacials. Additionally, authors conclude that cosmogenic nuclide concentrations present in modern sediments reflect that sediments are sourced primarily from incision of glacial deposits and not from bedrock erosion. Overall, I consider this manuscript to be written clearly, with data supportive of authors’ interpretations, with potential implications for future applications of this field and analytical framework in similar settings. I do have some comments for the authors to consider for the publication of this manuscript:

Thank you for taking the time to review our work. 

Main comment:

I encourage the authors to adopt a rather speculative approach regarding their main conclusions, especially when indicating the persistence of the Laurentide Ice Sheet during interglacials. The authors could potentially lean more towards explaining how to partition the different agents responsible for the obtained 26Al/10Be ratios: exposure duration, minimal erosion, and nuclide decay, which could be further elaborated in section 6.3. While the different ratios along the transect provide evidence for minimal erosion and burial near the center of the ice dome, offshore data and modeling from LeBlanc et al. (2023) ties the story together. The authors could potentially model different burial scenarios with the new terrestrial samples obtained in this work and the work published by LeBlanc et al. (2023) to weigh in on exposure duration. Otherwise, I suggest revising parts of the abstract and conclusions to be more speculative about the results.

Thank you for your suggestions. Both you and another peer reviewer have encouraged us to rethink our conclusions. We will revise the ms, especially in our discussion and conclusion sections to focus on the sediments that we measured. Because the measured 26/10 ratio can be the result of any number of different scenarios, partitioning as suggested by the reviewer is not strictly possible however, we can speculate about different ways in which the ratios we measure could have come about including long term sediment storage and the presence of ice through interglacials.

Line by line minor comments:

L45: Add citation on the collapse of northern Canadian ice domes.

Will do.

L49: I wonder if inheritance here refers to inheritance of cosmogenic nuclides, as one would naturally assume, or the word reflects preservation of bedrock features in the landscape. Consider clarifying.

We will reword for clarification. 

L67: Consider reorganizing the Background section to improve its organization and clarity. As of right now, the section opens with an introductory motivation paragraph. Then, dives into cosmogenic nuclides application, followed by LIS deglacial history. After those sections, the authors go back to explain the use of cosmogenic nuclides as tracers of sediment sources, followed by a subsection on ice sheet erosivity. I suggest either combining some of these sections to make it more coherent or set them in the way the narrative flow is preserved. For example: start with cosmogenic nuclides and then dive into the deglaciation of the LIS to provide a smooth transition to section 3. Field site.

We will reorder the sub-sections so that LIS deglaciation history is last in the background to easily transition to the field site description. 

L77: The last few words of this paragraph are quite similar to the first sentence of the first paragraph of this section (2. Background) and appears repeated. Consider unifying these sentences for improving the reading flow.

We will revise wording and rework sentences. 

L123: Consider adding a citation at the end of this first sentence.

We will add a citation. 

L149: Since this last paragraph argues about changes of the LIS at finer time scales, I would suggest including Wickert et al. (2023) GRL citation to the list to further support the sentence about LIS meltwater discharge and subsequent feedbacks.

Thank you for the suggestion. 

L197: Consider adding Barth et al. (2019) PP to the references here.

We will add it in. Thank you for helping us expand our referenced literature. This is one of the main points a previous reviewer stressed the importance of. 

L204: Please consider rewording.

Yes, this sentence is a little choppy. We will reword it.

L209: Don’t see these rivers on the maps. Consider adding them to the figures or provide more spatial/geographical context.

This is an excellent suggestion. We can easily add this to figures 1 and 3. 

L216: I don’t see how this last paragraph is relevant to this study. The authors don’t come back to the vegetation or precipitation later in the manuscript. I would suggest that authors consider removing this last paragraph.

We believe that it is important for readers not familiar with the area to understand the geographic context especially in terms of post glacial sediment movement and sourcing and so we don’t believe it’s appropriate to remove this paragraph.

L228: Consider adding an approximate direction of the transect.

We will consider this; thank you for the suggestion.

L230: Remove ‘as well as contemporary river sediment’. Otherwise, it’s redundant with the sentence that comes immediately after.

We will remove this. 

L234: I would move this last sentence to the end of the first sentence of this section to strengthen the application of this method and integration between terrestrial and offshore sediments published by LeBlanc et al. (2023).

This is a helpful suggestion. It would also allow us to more clearly emphasize the connection between the measurements of  nuclide concentrations in marine IRD and our dual-isotope deglacial terrestrial data. 

L253: Figure 1: I would encourage the authors to improve this figure. To begin with, there are several areas of the maps that are cropped, making names of cities, countries, and sites hard to read. Secondly, the font size and colors for some sites are unclear. For instance, lakes are depicted as gray polygons, yet labels are also gray, impeding a good readability. Additionally, there is a ‘Text’ label next to the label of the Labrador Sea that is erroneous, so please remove it. Lastly, I would suggest adding scale and north arrow to maps.

We will redraft this figure responding to the reviewer’s suggestions.

L255: Figure 1 caption: The Quebec-Labrador Ice Dome is noted in the figure as Q-L Dome, but not explained on figure. Simply, I suggest adding (Q-L) here in the caption.

We will add this in the caption. 

Line 260: Figure 2: Would adding the nuclide concentrations and ratios be more informative here in each photo?

We will do this, especially since figure 3 only shows samples by concentration, not name. 

Line 260: Figure 2 caption: Include river name corresponding to (MC-03).

We will add this. 

Line 303: Maybe ‘last accessed’ instead of ‘constants’?

We performed the Holocene exposure correction in January of 2024. However, the constants used by the CRONUS calculator were reported as 2020-08-26. We included the constants so our work could be reproducible. However, we will clarify to the reader when we last accessed the calculator. 

L415: Figure 6 caption: replace ‘stream’ with ‘river’ to be consistent with the same term used across the text. Also, it would be worth explaining what was the rationale behind separating some samples into distal and proximal and how this was done regarding the sampled transect and the likely transient evolution of an ice dome during interglacial(s).

We will swap the terms out for consistency. And we will be sure to better explain to the reader how we sorted samples into distal and proximal groups. Another reviewer also pointed out this lack of explanation. 

L431: Which subglacial processes were modeled? Also replace ‘modeling’ with ‘modeled’.

This is not modelling that we performed but that of Melanson et al (2013) as we cited in the text. It seems odd for us to repeat technical details of their paper but if the editor thinks this is appropriate we will. In our view, the results are what matters to most readers not the details of the approach.

L436: It is quite challenging to find these sites on the figures provided if you are not necessarily familiar with the field site. I would suggest either providing general location references for these sites or add them to the maps on Figs 1 and 2.

Thank you for this insight. We will add location references in the text. 

L480: Missing Staiger et al. (2005) in the reference list.

We will correct this oversight. 

L515: ‘taken from outcrops’ here is confusing. Either delete or consider rewording to improve clarity.

We will change this to ‘bedrock’. 

L531: Please consider briefly explaining how you arrived at this conclusion.

We will add to this paragraph and elaborate on the conclusion. 

L538: Please consider providing more details about the mixing model and/or how the results were computed. I found that this section could use some supporting information (figures or table in the supplementary material, for example).

We will add more details about the sediment mixing model in the next version of the manuscript and will incorporate your suggestion of adding relevant material to the supplementary section. 

L544: Add citation of average deglacial age for field area. 

Noted. Will do.

L561: I would suggest the authors to be more speculative about this statement. See main comments above.

We will re-examine our language throughout the paper. See reply to main comment. 

Supplementary material: I found it quite helpful to see field photos. I would, however, suggest revising some of the notes/descriptions since some notes are not relevant. Additionally, elevation units need to be changed so they are reported in the same system for consistency purposes.

Thank you for going over the supplementary material. We will edit elevation units for consistency and edit the field notes as appropriate.

References:

-Missing some references in the text that are listed in the reference list:

Pico et al. (2018), Abe‐Ouchi et al. (2013), Dyke (2004), Goehring et al. (2010), Gregoire et al. (2018), Rasmussen et al. (2006), Rice et al. (2019).

Thank you for catching this. They should be in parentheticals throughout the paper. We will fix this. 

REPLY TO REVIEWER(S) 3: Rice, Ross, Paulen, & Kelley

Review of Cavnar et al., “In situ Cosmogenic 10Be and 26Al in Deglacial Sediment….”

Responses to reviewer comments are in italics below.

The following is the result of conversations between: Jessey Rice, Martin Ross, Roger Paulen, and Sam Kelley.

Cavnar et al. provide a novel approach to investigate the subglacial conditions throughout glaciation of the Quebec-Labrador sector of the Laurentide Ice Sheet via an investigation of cosmogenic nuclides in deglacial sediments. The manuscript is well-written, interesting to read, uses novel approaches, and provides some useful insights, however, is missing some key references to past subglacial dynamic work and misinterprets some important previous research. Additionally, the conclusions of the manuscript are a bit too broad and slightly overreach given the data provided in the manuscript. The work does highlight some interesting findings that suggest further work is needed in this glaciologically important region.

We thank reviewers for their insights and will revise to include additional references. Comments from all three reviewers have led us to rethink our approach to discussing the data. In revision we will explicitly consider different ways in which the sediment we sampled could have the high concentrations of measured nuclides and in some cases, low 26/10 ratios.

We think the manuscript needs some improvements; some specific recommendations are suggested below:


Thank you for talking the time to provide this level of detail. It will be helpful in revision.

Line 47-49: In Rice et al. (2019, 2020), we did not find evidence of relict preglacial landscapes preserved due to sustained cold-based conditions, and we also did not find clear evidence that the oldest glacial bed fragments are from an older glaciation. We interpreted the older flowset (i.e., our Flow 1) as having formed during the last glaciation but likely before LGM.  In this work, through multiple proxies, we indicated there was no evidence of sustained cold-based conditions, but for this study in particular, from the limited 10Be samples we collected, we had no evidence to suggest high degree of inheritance from long exposures from previous interglaciations (when compared to other regions of 10Be with sustained cold-based conditions (Staiger et al., 2006; Refsnider and Miller, 2010; Corbett et al., 2016; Dubé-Loubert et al., 2021)). If you have any additional questions regarding this work, I would be happy to chat with you about it (jessey.rice@nrcan-rncan.gc.ca).

The author team read Rice et al., 2019 and Rice et al., 2020 and did interpret your findings as outlined above in your comment. Our use of ‘inheritance’ in line 49 refers to older inherited glacial landscape features, not inherited cosmogenic 10Be. In retrospect, this was a poor choice of wording because it was ambiguous. A previous reviewer pointed out the confusing use of ‘inheritance’ in this sentence as well. We will reword accordingly on the next draft of the manuscript. What is most important to acknowledge and we will do so in revision is that the presence of flow features is diagnostic of either warm based or polythermal ice over time and space.

Additionally, given the study area’s location, it would not be considered far northern Canada.

Noted, we will change this. 

Line 68-78: While we agree there is still little known about the pre-LGM conditions, there is work that suggests that most areas under the LIS in northern Quebec and Labrador were erosive at different times, possibly due to the migration of the Q-L. It is possible that cold-based conditions existed under the Q-L sector for some time, the divides were also moving across the vast region that it covered resulting in only transient cold-based conditions, with warm-based conditions also migrating across the sector as ice sheet dynamics evolved throughout glaciation.

This is a good point. The modeling literature we cite estimated variables including  ice velocity and basal sliding distance by time-integrating over the whole last glacial cycle. While it is important that this integration yields very slow moving, low erosive ice over northern Q-L, it cannot accurately depict transient cold based or warm based conditions that occurred in the region over the course of the last glaciation. As you point out, it also cannot represent how warm based or cold based conditions migrate spatially over time. We will add content to this section that acknowledges these model limitations. 

Although you cite the compilation works for this area (Kleman et al., 2009; Dalton et al., 2020), there are specific references used in those manuscripts that describe the cross-cutting relationship of landforms and striations that suggests continued warm-based erosive/deformational conditions at the base of the ice sheet (Veillette et al. 1999; Clark et al., 2000; Rice et al., 2019 (this one is listed in the references, but not cited in the text)).

Agree. What really matters is not whether the ice is warm or cold based but that the sub glacial erosion rates are low enough that nuclides from prior periods of exposure are preserved in bedrock and sediment because concentrations measured in both (by us and by others) exceed what would be produced since deglaciation. We will refocus our discussion to this salient point and away from thermal status of the bed, clearly acknowledging the prior work cited in this review and placing it in proper context.

While we also agree that some numerical models show low probabilities of warm-based conditions for this area, it is important to acknowledge their own uncertainty. Tarasov and Peltier (2007) highlighted the importance of using appropriate bed thermal conductivity and thus recognized the limitations of using a single value for an entire ice sheet model. Pickler et al. (2016- Climate of the Past) is one of a series of studies which analyzed deep borehole temperature profiles to reconstruct basal temperatures of the Laurentide Ice Sheet during the last glaciation at different sites. One of their study sites is close to your sample transect (Sept-Iles). In this study, they obtained temperatures near the pressure melting point at LGM, which was the coldest at all the sites analyzed. Interestingly, reconstructed pre-LGM temperatures were warmer for all the study sites including Sept-Iles. All this to say, determining bed conditions under all the LIS throughout the last glaciation is a difficult problem, and considerable uncertainty persists, but there is limited supporting evidence for widespread spatiotemporally stable and sustained cold-based conditions for most of the region of northern Quebec and Labrador. The evidence is patchy and uncertain at best and suggests regional variations and transient conditions. Most previous researchers interpret the glacial landscape of that large Q-L region as having formed during the last glaciation, suggesting broad warm-based conditions at times, which should be cited to present a more balanced representation of existing literature and interpretations. Similar evidence of these dynamic conditions has also been documented for the Keewatin sector in northern Manitoba and mainland Nunavut. 

We agree and will modify the manuscript accordingly as noted in response to the comment above. Upon your recommendation, we examined Picker et al. (2016-Climate of the Past). As mentioned above, the Sept-Iles borehole site is adjacent to our sample locations and had the minimum recorded temperature of -1.4°C out of the 13 sites (Picker et al., 2016). While this value is near the pressure melting point, the Sept-Iles basal temp was noted to be positioned on the ice sheet margin with thin, fast sliding ice (Rolandone et al., 2003; Margold et al., 2018). Empirical reconstructions based on 14C data and ensemble mean modeling place Sept-Iles at the site of a marine terminating ice stream that functioned throughout the majority of the last glacial period (Margold et al., 2018). No ice streams were modeled over our sample locations, further inland and north of Sept-Iles. This provides more evidence of strong regional variations within Quebec and Labrador—leading the author team to believe that it may be possible for some of our sample sites to have experienced colder bed conditions while adjacent areas were warmer and more erosive. There is really no way to know definitively which is why based on this and other reviews we have rethought our discussion approach.

Line 75-77: The relative timing of cross-cutting striations can be ascertained, but the absolute age of each flow cannot (i.e., Clark et al. 2000), although we think it can be constrained.  When erosional proxies are placed in the context of relative ice-flow chronologies, the erosive extent of ice flows can be evaluated (see Rice et al., 2019 and 2020). From this work, we indicated that Flow 1 was erosive and able to transport sediment significant distances (>100 km) and correlated this to Veillette et al. (1999) and Klassen and Thompson (1993)'s oldest flow and sourced from the St. Lawrence Highlands. This flow phase must have occurred pre-LGM as the flow was to the NE, and not radiating from an ice center, which would be expected at or near LGM.  

If this material remains in the revised ms, we will add some information to this paragraph clarifying that the relative chronology of flows can be determined (and cite some of your sources provided). In the end, it may become less relevant as we refocus the discussion.

Line 125-127: Arguably, the retreat of the LIS in northeastern Quebec is one of the most poorly constrained across the entirety of the LIS. It may be well constrained in southeastern Quebec/southwestern Labrador, but northeastern Quebec/western Labrador is poorly constrained north of 54° (basically Smallwood Reservoir over to Lake Melville). This should be rephrased.

We will be more specific when rephrasing this.  

Line 142-143: We would argue that most of the evidence now suggests Hudson Bay was not ice-free during MIS 3.  Hodder et al. (2023) provided evidence that there was not an ice-free Hudson Bay during MIS 3. See also Hodder et al. (2024) and Gauthier et al. (2024).

Thank you for recommending these sources. We will incorporate them into this paragraph and edit to accurately reflect what recent evidence suggests about MIS 3. 

Line 175-176: Rice et al. (2020) did find this as well, but further north and closer to a major ice divide of the Q-L sector, although found it was more a result of ice divide formation and migration than proximity to the centre of the dome. We found that inheritance across the study area was low, with only a few areas (that experienced ice-divide migration across them) having evidence of sluggish ice conditions and inheritance. A vast majority of the study area indicated little to no inheritance, suggesting significant erosion.

We are unsure if the review means landform inheritance here or cosmogenic – usage is uncertain. We will add Rice et al. (2020) to the parenthetical references used in this section as evidence of variable erosion by the Q-L region of the LIS. Similar to other literature that measured 10Be in boulders or erratics in Q-L, inheritance was present in a few samples but not the majority (Ullman et al, 2016; Couette et al, 2023; Rice et al., 2019; Rice et al., 2020). We can edit to make these similarities more explicit to the reader. Unlike these studies, all of our deglacial sediment samples had significant inherited concentrations of 10Be and 26Al. We are considering further developing this dichotomy in the manuscript—as we did not directly address possible explanations why nuclide inheritance is so common in deglacial sediments in our field area. The inheritance could imply that the ice sheet does not efficiently remove sediment even if it is eroding a meter or two of rock with each glacial cycle. Such sediment recycling could explain our observations and will be added to the discussion.

Line 200: and study site information: Given that the article focuses on glacial erosion, we think at least a brief summary of the glacial history of the study area is relevant and required. The soil thickness has little to do with the study's focus, the till and deglacial sediments are critical and their formation should be discussed.

Section 2.2 of the background provides a history of Q-L (and the LIS as a whole). However, a previous reviewer suggested that we move the paragraphs on glacial history to the end of the background section as a smooth transition to the study site description. We plan on making this change as well as expanding the last paragraph of this section to give the reader more information on Q-L regional glacial history. 

Soil and deglacial sediment as well as the soils that formed on them will become increasingly important as we diversify the discussion and consider others ways in which 26/10 ratio was depressed in sediment why deglacial sediment has such high nuclide concentrations.

Line 233-234: Why was only a single bedrock sample collected over this 1000 km sector? The reasoning for the single bedrock sample or the insights from this sample are not clear throughout the entire manuscript, it would help the reader understand the methodology better if this was explained.

Our goal was to collect many more bedrock samples. However, bedrock samples were collected ad hoc according to proximity to our route on the Trans-Labrador Highway, and we unfortunately only came across two within short distance from the road. One of them had an insufficient quartz and could not be used for 10Be and 26Al analysis. We also intended for this project to have two field seasons, but the second season (intended for summer of 2023) was canceled due to the Canadian wildfires, leaving us with n=1 bedrock sample. We could leave it out but seems a pity to strand data. 

Line 239- 240: We don't think this is an accurate representation of what Ullman et al. (2016) indicated. We are assuming this was pulled from their figure showing the extent of ice at ~ 7.6 ka? Do you mean it was the centre of the Q-L sector at that time?

Thank you for bringing attention to this. Yes, we pulled this from Ullman et al.'s figure. We will clarify that the Q-L Dome separated at ~7.6 ka into two 'sub-domes' (according to Dyke (2004), and that this was the center of the southern sub-dome at that time. 

253: FIGURE 1 The centre of the Q-L dome does not align with any of the portrayed isochrons, the timing of Ullman (2016)’s centre should at least be represented, or an explanation of how the centre was determined should be given. Also, a DEM would really help this figure and there should be latitude and longitude on the figures and a scale.  

We will edit the figure and add lat/long, a scale, and a DEM. Adding the ~7.6 ka estimated center from Ullman is a helpful suggestion. See also the reply to the above comment. 

339: FIGURE 3 There should be errors included with the measurements, it might help to have different shapes to convey the sample type (modern vs deglacial) with different colour fonts to make it easier to separate the two types of samples. As it stands, it is a bit difficult to ascertain which values correspond to what sample at the current figure’s resolution. This would help the reader to evaluate the data more easily.

Thank you for suggesting different colored fonts for different sample types. This will definitely make it easier for readers to correlate measurements with sample points. We feel that including the error with the measurements (can be found in Table 6) would crowd the figure. However, we will add a reference to Table 6 in the caption for Figure 3 so readers know where to look for specific error values. We will try redrafting this figure and see if there’s a way to add uncertainties.

405: How are you determining initial exposure? Could the sediment have been deposited under a thick till sheet and then eroded and incorporated into a deglacial sediments during a subsequent glacial event? This is very difficult to determine, but it should be discussed as a potential.

Yes, agree and we will do this in our revised discussion.

409-411: Could the source of those sediments not have been from anywhere along the eastern coast of Labrador? Why would it need to come from Hudson Bay? There is lots of evidence of sustained cold-based conditions in the Torngat Mountains (Staiger et al., 2006) and as far inland as the George River, QC (Dubé-Loubert, 2021).

We suggest on lines 409-411 that LeBlanc et al's sediments could have come from Hudson Bay or from deeply buried sediments elsewhere (not specified), but not that they necessarily need to come from Hudson Bay. But the reviewers make a good point and we will add that the eastern coast of Labrador could also be a possibility. 

FIGURE 6. There is no y-axis unit, difficult to tell the degree of probability for each plot, could you please add?

We will add this, thank you, 

421- 422: This is a very bold far-sweeping statement for the data provided. We would suggest saying that there was inheritance from a period of prior exposure, but extrapolating beyond that seems a bit too much for the data presented.

Agree, all we can know is that the concentration of cosmogenic nuclides in these samples represents prior exposure.

441-442: If you look at the surficial map of Canada (the best resolution for this statement) most of the area is covered in glacial sediments. Therefore, stating that sediment coverage was patchy and thin is not entirely accurate for the study area.

Attached as a supplement is a screenshot of the Pelletier et al. (2016) map we cite from http://dx.doi.org/10.3334/ORNLDAAC/1304. We think “patchy and thin” is a decent way to describe the Q-L region in comparison to much of the rest of the areas formerly covered by the LIS. But we can soften the wording a bit or quantify more to clarify. 

456-457: Given all of the variables and the difficulty estimating them (i.e., interglacial exposure duration) can you be sure the concentrations purely reflect erosion rates?

No they do not, they represent an impossible to decipher mix of exposure, erosion and decay and that will be the basis of our revised discussion.

460- 463: Minnesota is quite far from the Keewatin ice dome and is better described as the southern LIS margin.

We will clarify this, it is in fact both.

475-482: This review of individual papers allows for different methods of identifying an outlier. Why not use the same approach as was used in your determination of Holocene exposure history? Recalculate all the cosmogenic exposure ages from the region (including those from a wider range of papers and use a set of isochrons to determine which ages contain inheritance).

We will do this and plot all of these on a map for a new synthetic figure. Thanks for the suggestion.

511-516: There are certainly many places the quartz could have been sourced from, but there is significant ice streaming off the coast of Labrador inputting significant sediment into the North Atlantic, why couldn’t these quartz grains have come from somewhere along the Labrador coast?

This is a good point. The coast of Labrador isn't technically part of the region we sampled. We will add it as another possibility. 

544-545: This is minor, but is decadal resolution realistic for estimating deglaciation across such a vast area, or even a given area? (9.32 ka). It is understandable that this was derived from published isochron data, but removing the last digit is probably a better representation of the published data.

Noted. We will change this rounding to 100s of years.

561-562: We do not think this conclusion is valid based on the evidence presented, evidence of polythermal conditions, yes, but using this as evidence of continued glaciation over such a broad region is overreaching.

Our plan after receiving three reviews is to rethink and rework the discussion to focus on the sediment data and then proposed various scenarios that are consilient with measured isotope concentrations and ratios citing appropriate literature to support and refute different possibilities. Polythermal ice is one of those possibilities as is extended sediment storage and remnant ice masses.