General comments

This is a well-crafted paper that presents new insights on the deglacial chronology of a vast area of the NW LIS and is constrained by a significant number of new surface exposure dates. The paper addresses relevant scientific questions within the scope of TC and is of high scientific quality, and the results are well supported using thorough methods and modelling. Substantial conclusions are reached – namely the authors propose meaningful modifications to ice margin retreat positions in an area clearly lacking dated material, and the results have a significant impact on ice retreat dynamics, the timing of ice-saddle collapse, meltwater delivery to the Arctic Ocean and contribution to sea level rise, and on the idea of an early opening of the IFC for peopling of North America. 

The authors give proper credit to related work and clearly indicate their own original contribution. Model outputs and inputs, sensitivity analysis results for different GIA models, and description of model calculations provided in the text and Supplementary materials are complete and precise to allow their reproduction by fellow scientists. The overall presentation is well structured and clear; language is fluent and precise. The number and quality of figures is appropriate. Figure 6C is missing however. The references are generally appropriate. Several references cited in the text or in the figures are missing in the list, and some listed references do not appear in the text.

Overall a very good contribution that requires minor revisions outlined in the comments below.  Specifically minor parts of the text and figures should be clarified and/or modified. Technical corrections are required or highly recommended.

Specific comments:

45: Clarify timing of persistent IFC from the earlier models. During entire last glaciation?

73: Boulders suited for TCN analysis should be large in size (>1 m) and stable. Some boulders from this study appear rather small and/or considerably embedded in sediments on Figure S2. Explain reasoning for choosing such small boulders and if exposure ages may have been affected by movement or exhumation – in methods section or caption of Figure S2.

134-135: Provide range of % change of the minimal impact of changes in atmospheric mass distribution on the exposure age from cited studies.

299 and Fig. 8: The two clusters of ages at the Cap Mountain summit site are not obvious on Fig. 8 and S6. They appear more like a spread. Clarify.

308: Is the change in the retreat pattern and style provided by the new TCN ages also supported by ice marginal landforms and/or changes in the glacial erosional record?

335: Provide reference for given radiocarbon dates (13.4 ± 0.17 cal ka BP).

447: It is difficult to see the ice lowering of 116-157 m versus 19-47 m on Fig. 5A. Can you point to these drops on the figure?

478-481: Consider adding underlined text for clarification: … is an earlier retreat of the NW LIS by about 1000 years around 63N in the central Mackenzie Mountains...

Figure 2: Outline of Fig. 7 represents Fig. 9; clarify if ages besides the red outlines represent mean values of various samples; grey lines difficult to differentiate from drainage; also where does the LGM (21.1 cal) limit appear – difference with “local LGM limit at 18.0 cal”?; consider adding “(two sigma range)” after “deglaciation ages”.

Figure 3: Consider adding reference for ArcticDEM from Polar Geospatial Center.

Figure 4: Consider indicating location of Mackenzie Valley or River and locations of 6 studied sites.

Figure 5: Mention what are the red dots as well.

Figure 6: The 10Be versus elevation plot is entirely missing in C; add what are B-A and YD and the two vertical blue dashed lines.

Figure 7: Replace “points” by “lines” in caption of (A).

Figure 8: Delete one of the repeated “using the modelled age distribution output of” in caption.

Supplementary Document #1

- 14.2 cal ka: In fact, the ice margin north of ~65°N does not replicate that of Dalton et al. (2020). Explain the slight changes proposed.

- 13.5 cal ka: What is the basis for the changes of the ice margin position north of 65°N?  Geomorphology, elevation, TCN dates?

Technical corrections:

46-47: In addition to what? There needs to be a link to previous sentence here.

60: Although Consider replacing by “However”.

62: Reference should be Dyke et al 2003 (as in list).

64: Consider adding “(Fig. 2)” after 65˚N.

137: effect

142: Missing Jones et al., 2017 in reference list. Unless it should be Jones et al. 2019.

162-163: Missing Small et al., 2020 in reference list.

184: Consider referring to Table S2 at the end of sentence.

186: Should be Table S2 and not S3 here.

190: Consider adding “at the Summer Island site” after located.

201: Consider adding “we” in …and so we do not refer…

211: Move bracket before “see” (see Gregoire et al., 2016)

284: Add “was” after Mackenzie Valley floor.

305: Add reference to Fig. 2 at end of sentence.

306: Figure 4 does not appear in order as cited in the text. First time this figure is mentioned.

388-89-90: Rephrase – difficult to grasp.

393: Consider adding “(Fig. 6)” at the end of the sentence (2016).

453: Add “in” after deglaciation.

460: Missing (Menounos et al., 2017; Corbett et al., 2019) references in list.

463: Missing Ivanovic et al 2017 and 2018 references in list.

465: in

521-22: Delete repeated reference.

528-29: Reference not found in text.

562: After Barbett, P.J., consider adding “and others”.

622-23: Reference not found in text.

660-663: Reference not found in text.

708-709: Delete repeated reference.

739: Incomplete reference?

Table 1: Missing Berto et al (2022) reference in list.

Figure S1: Add references for each named model in caption: Peltier et al., 2015; Lambeck et al., 2017; Gowan et al., 2021.

Figure S2: Provide size of chisel(s).

Figure S3: Borchers et al., 2015 is given as 2016 in list of references. Verify.

Table S1: Provide Peltier’s reference instead of ICE-6G in the headers (to be consistent with others); add “age” to all headers.

The paper by Stoker et al. addresses the timing of continental glaciation in northwestern Canada during the last deglacial period using surface exposure dating and modeling results.

Overall, I find this paper to be well written and provides a fairly comprehensive assessment of the glacial history, prior data, and prior work on the ice history in this region.  The data itself are a nice contribution to the existing data and I think warranted for publication.  To me, the authors have done a nice job in both presenting their new work and putting it into the context of existing datasets.  Please see my line comments for mostly minor suggestions.  Some of the questions I have regarding the Bayesian modeling are a little more major as are the reporting of information regarding the 10Be ages, calculations, and lab procedures.

Line Comments:

Line 53:  Early 36Cl dates - how many dates were presented in the earlier work? Perhaps say.

Line 59: “On the all-time” – this is colloquial and hard to understand what is meant.  Consider rephrasing.

Line 65: “Dipstick” – define this for the reader.

Line 79: “True” – what is meant by true?  Can you define this better for the reader?

Line 98: HF – include the percentage since presumably not concentrated. 

Line 103: The carrier concentration needs to be included for replication purposes.  Very important.

Line 112: The blank values need to be stated.  I suggest included them in both the text and supplemental to help the reader understand the precision of the measurements.

Lines 117-118: Again, what are all these uncertainties?  Please state them (%) to help the reader understand how each is incorporated in the measurements.

Line 134: What did Cuzzone et al. find?  State what is meant by “minimal”.

Line 135: Model resolution – what is meant by suitable resolution?  The models that were used in Cuzzone et al. will have the same resolution for this location.  I don’t think resolution should factor into any of this or be a justifiable reason to not do it.  There is also statistically downscaled data for North America that could be used, so it might in fact be better than the model used in Cuzzone et al.

Lines 164-165: How does the spatial scale factor into the Bayesian model and the priors?  It is clear when doing an elevational transect with the 10Be ages but how do the authors account for the 14C ages coming from different locations.  Elevation and simple horizontal retreat of the margin will control the timing and it isn’t clear to me how it all fits together.  I realize this is challenging to explain but I do think it worth trying to explain this to the reader.  The supplemental Figures 7 and 8 are just the line code from Oxcal which really isn’t that helpful for parsing this all out.  At the moment, I don’t think a reader could reconstruct what the authors have done and/or the assumptions being made.

Line 158: General comment on the Bayesian modeling – what are the uncertainties being used in the model for both the 10Be and 14C ages?  Presumably, the authors are using the external uncertainties for the 10Be but this isn’t made clear in this section. More information would be useful.

Line 299 and Figure S2: Many of the boulders selected for this study are fairly small and several look like they are being exhumed.  The authors should discuss this either in the methods or results about the potential implications.  Or, alternatively, give the reader some sense if this is problematic or not (e.g. boulder height plotted versus age by location or in general).

Figure S2: It might be better to break these photos into locations either as new figures or into rows.

Line 275: I find this statement a little circular.  It assumes the regional deglaciation a priori and then uses it to argue for relatively minor affects.  40km is quite large when considering it in the context of the elevational sampling being done – one may even argue that the elevation changes are quite small themselves, thus top-down elevation dates could in fact be synchronous or even out of order within the uncertainties.  I think the authors need to better justify why this is in fact a minor issue – the surface slope of the ice may be a good place to argue and/or the modeling.

Line 282: What does vetted mean?  Try to be more exact and descriptive for your reader.

Lines 358-359: I don’t recommend phrasing things this way unless the authors have made some considerable effort reaching out to L. Tarasov or the authors of the Reyes et al. paper.  Did the authors in fact contact the authors?  Otherwise, it is just calling them in published form.  Please reconsider.

Lines 370-372: While I don’t disagree with the authors argument about the Arctic data and it being appropriate (or not).  I do think they need to justify things slightly better – the 47 samples may fit the elevation and ages ranges but the spatial changes in the production rate can be quite high because for a variety of reason (e.g. geomagnetic, snow, atmospheric, etc. affects).  It would be useful if the authors provided some sense of where these sites are located relative to their own site.  Clearly the Reyes et al. paper used the Arctic sites for this very reason so it would help the reader to understand the why the choice for this paper was made as clear as possible.

Line 460: Corbett et al. 2019 reference is missing.  Also, there are several other papers that could be cited here including Koester et al. and Barth et al. from the New England areas where dipsticks have been used to determine the timing of glacial retreat and presumably sea level contributions.

References in general: many references are missing.  Please do a thorough review to make sure you have all of them and all extra ones are removed.

Line 488: Data should never be made available on request.  Please provide all data in free, online repository following FAIR and journal data standards.

Figure 2: Make the text bigger in the figure – hard to read for us oldies.  Try to make clear what is new data from this study and what is existing data from others.  Some of the radiocarbon ages say “0.0 and null” for the uncertainty.    Is this correct and if it is 0.0 make it another decimal and if null, you need to explain what this means.

Figure 3: Are these 14C or 10Be ages?  If the latter, which I know they are, make sure to keep the symbols the same as you have in Figure 1.  Text is hard to read – consider making larger.  It’s also hard to know what is transect or not – can you make an inset for the elevational transects that demonstrates the ages with elevation?  This would help the reader understand the spatial and elevation data.

Figure 4: Need some reference to where the data were collected.  Maybe plot the points where the ages are from on the map.

Figure 5: Label the ages on the surface elevation profiles for the ice sheet.  This will help the reader out a bit.

Figure 6: Part C is missing from the figure.  I am somewhat confused on final results of the Bayesian modeling.  It seems to have picked the most improbable sections of the ages at each of the sites.  I think this is largely related to the primary assumption that the dates need to be younger with lower elevation.  Or, alternatively, the ages are too uncertain or there are inheritance/exhumation issues with the data to correctly apply this assumption for the Bayesian modeling effort.  As reported in this figure (e.g. part A), it would mean that the age tails are the most probable age for four of the PDFs which seems very unlikely to me.  This should be addressed.  Perhaps this gets clarified once we can see figure C.

Figure 9: I’m not sure if this figure is relevant to the main paper.

Table: the authors should provide a data table for input into Cronus (Balco et al. 2008).  This will allow the reader to easily replicate their work and if production rates or scaling change, easily adjust those data into the future.  This should be standard policy for all work using this or any calculator.  The exact input files should be provided.  Apologies if I overlooked it.