General Comments

This manuscript essentially summarizes more than two decades of research on stratified ice-wedge-filled lacustrine deposits indicative of interglacial warmth at two nearby sites in NE Siberia: Bol’shoy Lyakhovsky Island and the Oyogos Yar coast. Much of the biostratigraphic evidence has been published previously in specialty journals but is helpfully summarized and compared here, and includes important new geochronological evidence confirming an Eemian (MIS 5e) age and summarizes all of the biostratigraphic evidence in climate terms.  As such, this is a very useful synthesis for the Quaternary community and is well suited for Climate of the Past.  It provides an important synthesis of a massive amount of work, including a wider range of climate proxies than for almost any other LIG Arctic site. I am not especially familiar with these deposits, but there is quite an extensive literature on them.

                  An OSL expert should review those results as these are essential to the story and as near as I can tell, have not been previously published, whereas much of the other data have been published in specialty journals previously.  I have no reason to be suspicious, although the stated precision is somewhat better than most OSL ages in this time range.         There are an amazing amount of specific analytical results, all at least modestly useful, but certainly not of equal informative power. Still, the less significant results take up less space and document the breadth of effort put into these studies.

Paragraph indents would have been very helpful

Specific Comments

 Abstract  This should be a single paragraph that succinctly explains what is newly published in this paper and what is being summarized from previous (mostly proxy-specific) publications.  And ending with a summary of what the authors think are the key interpretations for Eemian climate (both winter and summer), how these compare with model-based reconstructions and what appears to be the important factor of the higher Eemian sea level, and ending with how what they have learned has relevance for predictions of future Arctic warming in an enhanced greenhouse world.   The current abstract is several paragraphs long and reads more like an Introduction than an Abstract.

Specific comments by page and line number

p.2 Abstract

Line 16  sedaDNA not sedaDNA; looks OK in main text

Line 11.  “new luminescence ages” Are the luminescence dates new with this publication? If so, then line 8 might best read “Here, we present new geochronological results and synthesize cryolithological,….”

Line 19:  “proxy data and also paleoclimate model results indicate a regional LIG climate significantly (ca. 5 to 10 °C) warmer than today”  What region? Maybe make this specific to high northern latitudes? 

p.3 Line 19  “The globally warmer-than-today Last Interglacial (LIG, ca. 130-115 ka)

Do we really know LIG is globally warmer than today?  The primary forcing (insolation) is limited to Northern Hemisphere summer and is actually negative for summer in the S Hemisphere.  Rising sea levels from NH ice sheets can destabilize some of Antarctica without warming. The Holocene appears to show no early Holocene warmth globally but strong early Holocene warmth in the Arctic…can Eemian be the same?  This needs a reference it the authors want LIG warmth to be global.

p.8 Line 20 and following            3.2 Luminescence dating

The section on Luminescence dating is important because it seems to be new results that confirm the age of the deposits to be indeed MIS 5e.  Please clarify when the sampling occurred and when the analyses were made and whether the dates cited came from only one of the two sites.  Were earlier efforts inconclusive?  Is this the first time these MIS-5e dates are being published?  Can you show a section where the OSL samples were taken and the context of other biostratigraphic samples in the same section.  This seems important to convince the reader that the dates have direct relevance to the climate reconstructions.

There are no citations in Section 4.2 Luminescence dating, hence I gather these results have not been previously published, and should be reviewed by an OSL expert.

p.10 Top

Pollen data are discussed in terms of processing, but no mention of how to deal with pollen from taxa with highly efficient wind-dispersal mechanisms. Particularly Alnus, Salix, and Betula that are very efficiently wind transported.  However, it appears that actual plant fragments of at least Betula and Alnus wee recovered.  I suggest presenting the plant macrofossil evidence first as its authenticity for on-site plant grow this much higher than for pollen, especially for taxa dependent on wind dispersal of their pollen.

Table 2 is very helpful in this regard

Also a discussion on page 37 addresses some of these issues

Fig 4 very helpful and convincing for ID of Eemian

p.12   Section 3.5 Clumped isotope analysis of biogenic carbonates and derivation of lake water δ18O

Wouldn’t this make more sense to read: Clumped isotope derived lakewater paleotemperatures  

p.30-32 5.2 Last Interglacial chronology and dating uncertainties

I’m not sure the summary of the range of ages available is essential here. Seems like focusing on luminescence techniques, as that is all that is presented for age control of these deposits. Other dated deposits are listed but as those results are not really discussed, I don’t see why they are relevant to the paper. Although Fig 11 is somewhat helpful even though not particularly relevant to the main thrust of this paper.

Section 5.3 is important and very helpful, as is Table 7

p.36, line 13  “farther vs further” “farther describes physical distance; further describes figurative distances”

p.38&39 Conclusions

This is the one paragraph that most “general readers” will look to.  Page 39 first paragraph discusses the temperature estimates from a range of proxies, especially warmest month.  But it gets a bit muddled on exactly “how much warmer than present day”, or pre-industrial, summer temperature estimates they are. It would be very helpful to have a better presentation of

• Recorded summer temperatures (or estimated pre-recent-warming warmest month temperatures
• The range of LIG estimated warmest month temperatures for the various proxies and an attempt to summarize how these might be compared to contemporary measured air or lakewater temperatures
• The modeled Eemian warmest month temperatures

And then the discussion of how a higher sea level during teh Eemian may in its own altered warmest month temperatures

This section is so important to the general reader that a bit more effort to distill all their amazing data into a comprehensive summary is important.

OVERVIEW

The Last Interglacial (LIG) is an important analog for current climate changes, but terrestrial LIG sites are scarce.  This manuscript provides a detailed review and paleoclimatic analysis of LIG deposits from lacustrine sediments found in permafrost settings at a series of sites in northern coastal Russia.  This is a major synthesis and update, building upon fieldwork conducted over the past 25 years and a remarkably wide-ranging suite of paleoclimatic and paleoecological proxies.  (Many prior papers have been published from these sites, this work also serves as a review and synthesis of these papers.)  Highlights include new age estimates, new paleoclimatic reconstructions from a wide variety of proxies, and comparisons of the proxy data to paleoclimatic simulations. 

This work is also highly valuable given the current geopolitical situation and difficulty of conducting fieldwork in Russia.  For at least the next several years, these observations are essentially irreplaceable.

In some ways, this reads as more of a ‘Quaternary’ paper than a ‘paleoclimate’ paper, given that the paper includes extensive detail about the stratigraphic sections and ecological interpretations in addition to the paleoclimatic reconstructions.  All of the proxies presented provide useful information about local environments and ecosystems, but not all are directly relevant to past climatic interpretations.  This paper is very long and very detailed, but its discussion is thorough and all lines of evidence are carefully weighed and integrated.

I provide below a few comments on the individual proxy-based (and model-based reconstructions), some of which may require revisions to the analyses.  These are followed by minor line-by-line comments and see also the attached PDF for recommended edits to make minor grammatical fixes, shorten wording, and sharpen some phrasings. 

PROXY-BY-PROXY COMMENTS

Some of the climatic reconstructions are based on the Mutual Climate Range approach (for plants, insects), which is a fairly crude way to derive temperature estimates (Table 2, page 20).  But the MCR-based results agree well with each other and other approaches.

The pollen-based climatic reconstructions use a modern pollen training dataset that spans the Northern Hemisphere, including Eurasia and North America.  I recommend redoing these analyses to remove the North American samples, because they mostly represent different species with somewhat different climatic tolerances from their Eurasian counterparts. 

Additionally, the pollen-based climatic reconstructions also do not indicate whether a minimum SCD threshold was used to remove or flag no-analog samples, i.e. fossil samples likely to have no modern analogue.  This no-analog analysis and thresholding should be performed.

It’s exciting to see that sedaDNA analyses were performed at a core in this site, but it’s not clear what is added by these analyses.  No attempt is made to make a paleoclimatic inference based on the sedaDNA data (this is probably wise).  Grouping the aeDNA results into functional groups is reasonable (Figure 7, p. 21) but misses a lot of the interesting ecological detail.  Recommend adding a figure that shows the stratigraphic occurrences of the taxa described in the text on p21, at maximum feasible taxonomic resolution.

Similarly, the ostracode and mollusk analyses (pp12-13) are interesting but do not contribute to the paleoclimatic reconstructions shown here.  Delete or move to supplementary information?

Paleoclimatic simulations often include a spin-up period, during which the model is in a transient state that is artificially influenced by the starting conditions.  Was this spin-up period removed prior to calculating climatological means from the simulations?

Also, for the paleoclimatic simulations:  On P13, L23-27, I can’t quite follow this description of the calculation of the anomalies, which seems to repeat the subtraction used to calculate the anomalies, when it should only be performed once.

LINE-BY-LINE COMMENTS

P2L2-4:  Long sentence, awk

P2L31-33:  Model results are presented as anomalies but all proxy-based reconstructions earlier in this paragraph are presented as absolute values, making it difficult to compare the model-based and proxy-based results. 

P2L36:  Again, proxy-model mismatch is highlighted here, but the exact magnitude and nature of the mismatch is unclear.

P2L36:  What does ‘more systematic’ mean? 

P3L7:  ‘promote’ is present tense but prior sentences in past tense.  Check for verb tense consistency throughout ms.

P3L19-20:  Add ref(s) here to support LIG as analog for future, e.g. (Burke et al. 2018; Gulev et al. 2021; Otto-Bliesner et al. n.d., 2)

P3L26-27:  reduced ice sheet – reduced by how much?

P3L35:  A data-model mismatch is invoked here as if already introduced to the reader, but so far no data-model mismatch has been described.  Instead, the paper describes data-model agreement earlier in the paragraph.

P4L9:  ‘were conducted’… by who?  Unclear if this is introducing work by the authors or reviewing the broader literature.

P9L5:  Italicize Rumex and all other genus and species names.

P6 Figure 1:  Recommend replacing the 1-16 site labels with the site codes used in Figure 2.  Right now there is no easy way to look between Figures 1 & 2 and understand which sites are which.  Or, add the numeric codes in Fig. 1 to Fig. 2.

P7 Figure 2:  What is a ‘taberal’ deposit?

P16 Table 1:  A couple of the numbers are using commas instead periods to indicate decimal places.

P18L21:  ‘relatively’ – relative to what?

P19L2:  avoid ‘optimum’ when referring to a peak in temperature.  Any given temperature will be optimum for some species and suboptimal or adverse for others.

P22L6:  Is Morychus viridis a steppe species?  This is implied but unclear.

P22L8:  Which ‘other Pleistocene samples’ – those at the site, worldwide, or other domain?  Unclear.

P26L15:  Should A. Kossler be invited to be a co-author, given their intellectual contribution in identifying molluscs and the large interdisciplinary team?

P28L13-14:  agreement with each other?  Agreement with the proxy data?

P33-34, Table 7: 

*Recommend moving the ERA reanalysis data to the top row for each region, so that this modern reference is clearly established as the basis of comparison for all the paleoestimates. 

*Also note that the ERA reference is for 1990-2019, which already includes >1-2C of anthropogenic warming.  Consider also adding a row with a temperature estimate for the early 20^th century or pre-industrial period.

*For numbers with a +/- uncertainty estimate, does this represent one or two standard deviations?  Clarify in table legend.

P34, Fig. 12:  Recommend a standard order of first showing the modern values, then the paleoclimatic simulations, then the proxy results.  This ordering will better clarify the distinction between paleoclimatic simulations vs. proxy data.  Among the proxy results, I suggest moving the clumped isotopes to the bottom of the order, because they represent different variables than the other proxies.

P37L7-10:  Another reason why pollen is unlikely to be an important source of aeDNA is a mass effect:  the total biomass of airborne pollen rain is very low to all local aeDNA sources, so any pollen-sourced aeDNA is swamped out.  

REFERENCES CITED

Burke, K. D., M. Chandler, A. M. Haywood, D. J. Lunt, B. L. Otto-Bliesner, and J. W. Williams. 2018. “Pliocene and Eocene Provide Best Analogues for Near-Future Climates.” Proceedings of the National Academy of Sciences 115: 13288–93. doi:https://doi.org/10.1073/pnas.1809600115.

Gulev, S. K., P. W. Thorne, J. Ahn, F. J. Dentener, C. M. Domingues, S. Gerland, D. Gong, et al. 2021. “Changing State of the Climate System” eds. V. Masson-Delmotte, P. Zhai, A. Pirani, S. L. Connors, C. Pean, S. Berger, N. Caud, et al. Climate Change 2021: The Physical Science Basis. Contribution of Working Gorup I to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change.

Otto-Bliesner, B. L., N. Rosenbloom, E. J. Stone, N. P. McKay, D. J. Lunt, Esther C. Brady, and J. T. Overpeck. “How Warm Was the Last Interglacial? New Model–Data Comparisons.” Proceedings of the Royal Society A 371: 20130097. doi:https://doi.org/10.1098/rsta.2013.0097.

Review of Lutz et al. Climate of the Past

https://doi.org/10.5194/cp-2024-74

Thank you to the editorial staff and the authors for the extra time for reviewing this paper.  My appreciation also goes to the authors and their collaborators for pursuing an archive like this exposed in collapsing coastal bluffs. These sites are very special in the arctic regions and especially important as we are cut off from Russian collaborations for the foreseeable future.

I am very excited about this paper overall,  but have some concerns about presentation, figures, and context. I hope my comments increase the value of the manuscript. The extension of treeline to the arctic coast in northeastern Arctic Russia during the MIS 5e was first discussed by Lozhkin, A. and Anderson, P.: The last interglaciation in northeast Siberia, Quaternary Res., 43, 147–158, 1995. and then in Lozhkin and Anderson, CP 2013. These papers are about sites to the east of this paper by Lutz but the coherence or contrast is important to the community understanding of the nature of 5e in these remote arctic settings of Arctic Russia. I think Lutz et al should mention this work in the conclusions.  Lets hope collaboration with Russian scientists continue but right now we need to celebrate and cite important collaborations that contribute to paleoclimate science in the Arctic.

My comments on the paper are documented by page and by line number.

Abstract page 2 - Line 15,  Clayish Silt is not used.  Should be clayey silt or clay-rich silt.

Page 3 - line 4 -- superimposed "on" local factors.

Page 4  -- move Fig 1 closer to where its first used here line 10

Page 6  -- figure 1 and Figure 1b and 1C have numbered locations that are totally different from the sections (e.g., L14-12) in figure 2. Also Figure 2 b should be flipped so that the east end of the coastline is in the east as shown in Figure 1C. Why not put the section IDs on Figure 1 and not the numbers?

Page 7 line 2 - eastern parts, Bol’shoy Lyakhovsky Island is shaped by hills reaching elevation of about 100-300 m. Could you include a LIDAR or shaded relief map? Its not shown in Fig. 1. Yedoma or yedoma uplands?

Page 8 line 28 -- something missing in this sentence. Do you want to say " K-feldspars with grain sizes of 40–63 µm were extracted for IRSL dating from core samples L14-12-OSL1, L14-12-OSL3), but coarser grains, 63–90 μm, were extracted from cores at L14-12-OSL1.

Page 8 lines 33-36. Confusing to follow. Extracted medium-sized feldspar grains were used to prepare sets of 24 aliquots with 2 mm diameter??  is 40-63 um medium and what is 2 mm referring to?

Page 9 - line 1 -- what is cut-heat temperatures?

Page 10 - This might need more explanation and it’s my lack of knowledge perhaps. Dot occurrences in this GBIF database. I had to go look that up.

Page 12 - nitrogen-cooled coldfingers? Part of the instrumentation?

Page 13-14 - The climate modeling was confusing, but I think you were using existing ensemble models  from the CMIP6 project, (needs a year) and extracting their data for your field area. So, you are not running the models again but extracting from published models? Is this correct?  This section might need a figure.  The issue you point out about trying to run the model without knowing what the land configuration was during 5e is a good one so you might want to bring Figure S3 up into the main paper, not the supplement.  Land Point needs to be capitalized in the figure caption as you have it through the paper.

Page 14- line 25 -- clay-rich silt, not clayish silt.

Page 16 -- line 14. -- TIC values are not shown in Figure 5a.  Typographic error somewhere?   should this be TOC?

Results section -- It was long to read but you do go step by step through the proxies and point out how some match and some don't. But that probably to be expected because of calibration errors for each.  Very through.

Page 24 -- line 3 -- add reference to Figure 2 for Profile R35.

Page 28 -- Figure 9 and S3 are interesting, and it shows the failure of not having a fine enough grid size.  Not your fault but of models and trying to work that this local scale.

Page 30 line 22-23 -- that first sentence is a mess. I am not sure how to even reword it. perhaps it needs to be 2 sentences?

Page 31 lines 25-34. Yes, water or ice would impede the dose rate. So, agree that all you can do is assume an uncertainty.  And it is good to point out the overburden issue, but I thought it was not an issue below 1-2 meters of the exposed surface. I am not surprised that the age differences between 35 vs. 29 m is tiny.

Table 7 page 33 and 34 -- hard to read. I suggest leaving all of the n/a boxes empty so that boxes with data stand out.

Page 36 -- around lines 15 and/or 27. Here is where you should cite the work in Chukotka by Lozkhin and Anderson at Lake El'gygytgyn and in an older synthesis the migration of treeline in 5e (QR paper).  Your work and theirs paints a very clear image of the near loss of tundra long the coast during that time.

Page 39 -- lines 10-15 -- very good points here for future work.

Page 39 -- Strong summary.