the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 07 Oct 2024
Sedimentary ancient DNA insights into foraminiferal diversity near the grounding line in the western Ross Sea, Antarctica
Abstract. Foraminifera are important marine environmental indicators widely used in paleoceanography and paleoclimate studies. They are a dominant component of meiobenthic communities around the Antarctic continental shelf, including rarely studied locations below the ice shelves, close to the grounding line. In this study, we use high-throughput sequencing of sedimentary ancient DNA (sedaDNA) targeting foraminifera with two molecular markers, including the ultra-short one newly designed for this study, in five cores from the western Ross Sea, containing sediments up to thirty thousand years old. No foraminiferal DNA is detected in the tills, suggesting a lack of preservation of sedaDNA during glacially induced sediment reworking and transport. We reconstruct diverse foraminiferal communities in the open marine settings and significantly less diverse communities in sediments from the slopes of the grounding zone wedges, deposited proximal to the grounding line. Both assemblages are rich in soft-walled monothalamids not preserved in the fossil record and complement the results of earlier micropaleontological studies, allowing a more complete reconstruction of past biodiversity. The newly designed mini-barcode provides higher foraminiferal diversity in surface and subsurface samples than the standard barcode and allows better differentiation between foraminiferal communities in different sediment types. It appears to have great potential for future paleoenvironmental studies, although its taxonomic resolution needs to be evaluated.
- Preprint
(1317 KB) - Metadata XML
-
Supplement
(242 KB) - BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2024-2824', Anonymous Referee #1, 10 Nov 2024
General Comments
This study explores the diversity and composition of foraminiferal communities in the Western Ross Sea, Antarctica, particularly near paleo-grounding lines—areas that are critical for understanding past Antarctic ice sheet dynamics. Using sedimentary ancient DNA (sedaDNA) sequencing, the researchers examined foraminiferal DNA from five sediment cores, dated to as far back as 30,000 years, collected from both glacier-proximal and open marine environments. This innovative approach allowed the detection of both calcareous foraminifera and soft-walled monothalamids, a group rarely preserved in fossil records due to their fragile structure. The study reveals notable differences in foraminiferal diversity across sediment types, with open marine sediments exhibiting substantially higher diversity than those near glacier-proximal settings, where community composition was more limited. Notably, the researchers designed a new mini-barcode to improve the detection of monothalamid foraminifera, enabling the identification of species previously undetectable by fossil-based methods. This mini-barcode improved taxonomic resolution between foraminiferal communities in different sediment types and holds promise for broader application in future paleoenvironmental studies, though it may require further refinement for enhanced accuracy. One major finding was the lack of foraminiferal DNA in glacial tills, indicating that sedimentary DNA may not preserve well in reworked glacial sediments. This finding implies that stable, open marine environments may be more conducive to DNA preservation, offering more reliable records in such settings. Overall, the study underscores the potential of sedaDNA for reconstructing past biodiversity, especially in environments where fossil records are incomplete or absent, thereby contributing new perspectives on the ecological history of Antarctic grounding line environments. This work offers valuable insights into Antarctic marine biodiversity, and the findings, particularly concerning the use of sedaDNA to identify previously undetected monothalamid foraminifera, could significantly enhance paleoenvironmental reconstructions in glacimarine settings. However, there are several areas where the manuscript could be improved for greater clarity, coherence, and scientific rigor. (I recommend major revisions primarily focused on restructuring the manuscript and clarifying the text, rather than on redoing analyses.)Specific Comments
Firstly, the manuscript would benefit from a more refined structure, as the introduction does not fully explain the rationale for certain methodological choices—specifically, the need for a shorter barcode, which later becomes a focus of the discussion. This omission creates a somewhat fragmented flow and shifts the balance of the manuscript, as significant introductory material about sedaDNA methodologies and barcode design only appears later in the discussion section. A more focused introduction that fully describes the study’s objectives would establish a clearer narrative, reinforcing the purpose of the methods and findings. In brief, it’s essential to introduce the main findings that are later discussed (and even appear in the abstract).
While the results section is accompanied by effective figures, the text itself relies heavily on subjective descriptions and lacks specific quantitative data. Terms like “somewhere in between” (L. 207) and “very small” (L. 237) are subjective and would be improved with precise values, enabling a clearer interpretation of findings. A more quantitative approach to presenting results would better support the study’s conclusions. Furthermore, the statistical methods, while mentioned in figure captions, are not adequately explained within the main text. Including detailed explanations of each statistical test, the conditions under which they were applied, and whether they were one- or two-sided, would enhance the transparency and reproducibility of the study’s analytical approach.
The methods section is generally well-constructed, but further elaboration on statistical analyses would strengthen the scientific rigor. Clear descriptions of each test and a justification for their selection would improve the robustness of the analysis. Expanding this section would also improve reproducibility, allowing future researchers to better follow the study’s procedures.
Lastly, the manuscript would benefit from a thorough proofreading by a native English speaker. At times, certain phrasing or wording makes it difficult to discern the authors' intended meaning. While such a review could be completed following major revisions, it would greatly improve the manuscript’s readability.
Technical Corrections
L30-31: The connection between Arctic warming and Antarctic ice sheets is not clearly articulated. It feels abrupt, with no clear link.
L37: "More advanced" is too vague—does it imply previous techniques were not advanced? Being from the sedaDNA field myself, I understand that you might mean “recently developed” or “innovative” but you need more care with your wording choice.
L49-51: Monothalamids are hard to track in fossil records, and sedaDNA provides an alternative means, but what role do monothalamids play in the ecosystem? Why are they of interest? A brief introduction to their ecological significance would provide context.
L61-63: There is no clear mention of the need for a new mini-barcode, which is emphasized in the abstract as a major result. Why was a shorter barcode necessary?
L65-68: I find this sentence difficult to follow. Consider rephrasing for clarity.
L70-73: The stated goals don’t mention developing a shorter marker, yet this is presented as a significant outcome in the abstract and discussion. Clarify if this is a primary aim or a secondary result.
L151: Unusual spacing between “and” and “3uL.”
L169: A fixed 90% coverage threshold affects the markers differently, as the shorter marker would allow more mismatches. Please clarify this point.
L195: The notation “135 80” is unclear—is this 13,580, 13 580, or two separate values?
L195-196: The sentence is unclear regarding replicates for all samples except those at 200 and 280 cm. Please clarify.
L207: “As significantly as” may suggest a statistical difference. Please ensure accurate wording to avoid misinterpretation.
L216: Does “they” refer to datasets or patterns? Clarify what is being described as “more scattered.”
L218: “Slightly different” is subjective. Aim for a more precise description (other instances of subjective terms were noted throughout the results but not all listed here—please review for accuracy).
L222: “Tend to form”—do all samples follow this pattern, or only some? Provide specific values.
L237: “Very small numbers” is vague—specify.
L260: "Highly irregular"—what is considered regular in this context? Constant over time or without variation?
L260-275: The description of SH and ST detection of OTUs is difficult to follow within this paragraph. Consider restructuring for clarity.
L280-304: This section lacks discussion or interpretation of results and instead offers a lengthy introduction to sedaDNA preservation.
L286: Unusual spacing between “sediments.” and “To determine.”
L310: When discussing sediment mixing and dilution, does this imply mechanical damage to DNA fragments or dilution due to sedimentation rate? Clarify.
L321: Could you report DNA concentrations post-extraction or gel bands after PCR amplification?
L323-326: Shotgun sequencing could confirm DNA preservation. Even without sequencing, a Bioanalyzer or TapeStation could verify fragment distribution, which may confirm the need for a shorter marker.
L377-409: This key discussion point about shorter markers was not introduced earlier—consider integrating into the introduction.
L378-382: The rationale for a shorter marker should appear in the introduction.
L404-405: Have you checked databases for potential sequence overlaps? Your taxonomic assignments should follow the Last Common Ancestor (LCA) principle. Therefore, the use of a shorter fragment would lead to potential lower taxonomic resolution (i.e., the inability to differentiate between two species) rather than misassignment. If all species within Borelis and Planoperculina are exclusively tropical and show no sequence overlap with the ST marker what are the closest relatives present in the databases?
L453: Unusual spacing between “(2023).” and “(1).”
L454-473: Conclusions should reflect the study's aims, with better alignment between introduction and final conclusions.Figures:
The figures are generally clear. You may wish to consider scaling the y-axis with depth, allowing for easier identification of core coverage. Also, note that only two dating points are provided for the entire record, which may limit chronological confidence.Citation: https://doi.org/10.5194/egusphere-2024-2824-RC1 -
AC1: 'Reply on RC1', Wojciech Majewski, 20 Nov 2024
Thank you for your comments! We will incorporate them into the revised manuscript, following further guidance from the editor.
Citation: https://doi.org/10.5194/egusphere-2024-2824-AC1 -
AC2: 'Reply on RC1', Wojciech Majewski, 18 Dec 2024
Thank you for reviewing our paper. The answers to the comments (in bold) are below. We do hope they are satisfactory.
Sincerely,
Wojciech Majewski, Ewa Demianiuk and co-authors
General Comments by Rev. 1.
Firstly, the manuscript would benefit from a more refined structure, as the introduction does not fully explain the rationale for certain methodological choices—specifically, the need for a shorter barcode, which later becomes a focus of the discussion. This omission creates a somewhat fragmented flow and shifts the balance of the manuscript, as significant introductory material about sedaDNA methodologies and barcode design only appears later in the discussion section. A more focused introduction that fully describes the study’s objectives would establish a clearer narrative, reinforcing the purpose of the methods and findings. In brief, it’s essential to introduce the main findings that are later discussed (and even appear in the abstract).Response: Thank you for pointing this out. We will explain the motivation for testing the shorter barcode in the introduction and add this as another aim of the study. At the beginning of our writing, we regarded it as more of a matter of technicality, but you are right, it is important to introduce it earlier.
While the results section is accompanied by effective figures, the text itself relies heavily on subjective descriptions and lacks specific quantitative data. Terms like “somewhere in between” (L. 207) and “very small” (L. 237) are subjective and would be improved with precise values, enabling a clearer interpretation of findings. A more quantitative approach to presenting results would better support the study’s conclusions.Response: Thank you, this will be corrected.
Furthermore, the statistical methods, while mentioned in figure captions, are not adequately explained within the main text. Including detailed explanations of each statistical test, the conditions under which they were applied, and whether they were one- or two-sided, would enhance the transparency and reproducibility of the study’s analytical approach.
The methods section is generally well-constructed, but further elaboration on statistical analyses would strengthen the scientific rigor. Clear descriptions of each test and a justification for their selection would improve the robustness of the analysis. Expanding this section would also improve reproducibility, allowing future researchers to better follow the study’s procedures.
Response: Thank you for the feedback. We agree that providing detailed explanations of the statistical methods in the M&M section would enhance the clarity and accessibility for readers. To address this, we will revise the main text to include detailed descriptions of each statistical test and related package.In the revised version. I will provide more details for statistics sections.
Lastly, the manuscript would benefit from a thorough proofreading by a native English speaker. At times, certain phrasing or wording makes it difficult to discern the authors' intended meaning. While such a review could be completed following major revisions, it would greatly improve the manuscript’s readabilityResponse: The final revised manuscript will be read and corrected by a native speaker.
Technical Corrections
L30-31: The connection between Arctic warming and Antarctic ice sheets is not clearly articulated. It feels abrupt, with no clear link.Response: Thank you. For simplicity, Arctic warming will not be mentioned here.
L37: "More advanced" is too vague—does it imply previous techniques were not advanced? Being from the sedaDNA field myself, I understand that you might mean “recently developed” or “innovative” but you need more care with your wording choice.Response: Yes, “recently developed” is a better choice here. Thank you.
L49-51: Monothalamids are hard to track in fossil records, and sedaDNA provides an alternative means, but what role do monothalamids play in the ecosystem? Why are they of interest? A brief introduction to their ecological significance would provide context.Response: Sentence “Monothalamid foraminifera are particularly well represented in marine restricted environments such as fjords, including environments close to the glacier fronts (Majewski, 2010; Korsun et al., 2023), which was also confirmed by metabarcoding analyses (Nguyen et al., 2023).” from lines 51-53 will be placed earlier on.
L61-63: There is no clear mention of the need for a new mini-barcode, which is emphasized in the abstract as a major result. Why was a shorter barcode necessary?
Response: Thank you, yes this will be added.L65-68: I find this sentence difficult to follow. Consider rephrasing for clarity.
Response: Is this roughly, ok? “Although attempts have been made to assess diversity of Antarctic foraminiferal using metabarcoding (Habura et al., 2004; Pawlowski et al., 2011; Li et al., 2023), this is the first time when Southern Ocean subsurface sediments has been targeted and one of only a few analyses of sub-surface sediments worldwide (Lejzerowicz et al., 2013; Pawłowska et al., 2014, 2016, 2020a, 2020b; Szczuciński et al., 2016).”
L70-73: The stated goals don’t mention developing a shorter marker, yet this is presented as a significant outcome in the abstract and discussion. Clarify if this is a primary aim or a secondary result.Response: As stated above, this will be added as one of the aims in the introduction.
L151: Unusual spacing between “and” and “3uL.”Response: This will be corrected.
L169: A fixed 90% coverage threshold affects the markers differently, as the shorter marker would allow more mismatches. Please clarify this point.
Response: A fixed 90% coverage threshold impacts markers differently because shorter markers inherently tolerate a greater proportion of mismatches or gaps while still meeting the 90% similarity criterion. The differential effect can influence the taxonomic resolution and accuracy of sequence assignments across markers of varying lengths. Based on the length of 2 markers, the 90% identity with allowing up to 7 mismatches or gaps and 90% coverage was fixed for BLAST to reduce the differential effect.
L195: The notation “135 80” is unclear—is this 13,580, 13 580, or two separate values?Response: The comma notation will be used throughout.
L195-196: The sentence is unclear regarding replicates for all samples except those at 200 and 280 cm. Please clarify.Response: The sentence “SH samples in KC30 from 200 and 280 cm were analysed in two replicates.” will be moved to the method section, where it fits better.
L207: “As significantly as” may suggest a statistical difference. Please ensure accurate wording to avoid misinterpretation.Response: Thank you. “Strongly” may be better then.
L216: Does “they” refer to datasets or patterns? Clarify what is being described as “more scattered.”Response: “They” will be changed to “data points”.
L218: “Slightly different” is subjective. Aim for a more precise description (other instances of subjective terms were noted throughout the results but not all listed here—please review for accuracy).Response: “clusters for surface and subsurface samples do not overlap”
L222: “Tend to form”—do all samples follow this pattern, or only some? Provide specific values.Response: “The surface open-marine samples from different cores form a single tight cluster.” Thank you it was a vague statement, indeed.
L237: “Very small numbers” is vague—specify.Response: “small numbers, i.e., < 3%”
L260: "Highly irregular"—what is considered regular in this context? Constant over time or without variation?Response: “with depth” will be added.
L260-275: The description of SH and ST detection of OTUs is difficult to follow within this paragraph. Consider restructuring for clarity.Response: Thank you, we will work on this paragraph.
L280-304: This section lacks discussion or interpretation of results and instead offers a lengthy introduction to sedaDNA preservation.Response: This part will be shortened and moved to the Introduction as the motivation for aiming to develop the shorter barcode.
L286: Unusual spacing between “sediments.” and “To determine.”Response: It will be corrected.
L310: When discussing sediment mixing and dilution, does this imply mechanical damage to DNA fragments or dilution due to sedimentation rate? Clarify.Response: Thank you, more of the first, i.e., mixing/reworking.
L321: Could you report DNA concentrations post-extraction or gel bands after PCR amplification?Response: We checked DNA concentration after extraction for each sample only to make sure that there is a genetic material for further analysis. Measurements were conducted with two different machines (one part of study were performed after long time), which is the reason we decided to not included the numbers.
L323-326: Shotgun sequencing could confirm DNA preservation. Even without sequencing, a Bioanalyzer or TapeStation could verify fragment distribution, which may confirm the need for a shorter marker.Response: Thank you, we will add this suggestion as one of the recommendations.
L377-409: This key discussion point about shorter markers was not introduced earlier—consider integrating into the introduction.Response: Yes. It will be done.
L378-382: The rationale for a shorter marker should appear in the introduction.Response: As above.
L404-405: Have you checked databases for potential sequence overlaps? Your taxonomic assignments should follow the Last Common Ancestor (LCA) principle. Therefore, the use of a shorter fragment would lead to potential lower taxonomic resolution (i.e., the inability to differentiate between two species) rather than misassignment. If all species within Borelis and Planoperculina are exclusively tropical and show no sequence overlap with the ST marker what are the closest relatives present in the databases?Response: Thank you for pointing this out. We agree with the reviewer's comment that SH provides a lower taxonomic resolution since no OTUs were assigned to tropical species in the ST. The assignment for both ST and SH markers was done using BLAST best hit as in M&M (90% similarity and 90% coverage and 7 gaps or mismatches). As a result, some SH sequences were assigned to tropical species. In the main text, we have provided and discussed criteria for avoiding misidentification.
L453: Unusual spacing between “(2023).” and “(1).”Response: It will be corrected.
L454-473: Conclusions should reflect the study's aims, with better alignment between introduction and final conclusions.Response: Yes, Introduction will be expanded.
Figures:
The figures are generally clear. You may wish to consider scaling the y-axis with depth, allowing for easier identification of core coverage. Also, note that only two dating points are provided for the entire record, which may limit chronological confidence.Response: Thank you. In Figs. 2, 6 and 7 “Depth (cm)” axes labels are missing. They will be added.
Citation: https://doi.org/10.5194/egusphere-2024-2824-AC2
-
AC1: 'Reply on RC1', Wojciech Majewski, 20 Nov 2024
-
RC2: 'Comment on egusphere-2024-2824', R. Mark Leckie, 30 Nov 2024
To be honest, the DNA science of this paper is out of my comfort zone, but I am very familiar with most of the protists (the foraminiferal species) described in this paper (although I'm not at all familiar with monothalamids), and I'm very familiar with forams of the Ross Sea and the Neogene glacial history of this part of Antarctica. This is truly a fascinating paper, which is clearly contributing to new and exciting ways to analyze fossil DNA, including the monothalamids which contribute significantly to the DNA signature of the sediments in the Ross Sea. The authors are using a new tool that is allowing them to probe more deeply into the nature of foraminiferal diversity in a variety of depositional environments ranging from open Ross Sea to sub-ice shelf to proximal grounding zone. I cannot comment specifically about the DNA and sequencing science, but I have made a few minor editorial suggestions for the authors, and I have raised some questions and have made a few observations based on recently published papers from my lab group that may have relevance to their findings. The results of Demianiuk et al. have certainly opened my eyes to new ways of uncovering the complexity of the glaciomarine environment and the microfossils that I study.
Great paper. Very minor edits from my perspective. Look forward to seeing this published.
Mark Leckie, University of Massachusetts Amherst
-
AC3: 'Reply on RC2', Wojciech Majewski, 18 Dec 2024
Thank you for reviewing our paper. All the minor corrections marked on the pdf will be implemented. The answers to larger comments (in bold) are below. We do hope they are satisfactory.
Sincerely,
Wojciech Majewski, Ewa Demianiuk and co-authors
L44: Seidenstein et al. 2024; Marine Micropalaeontology
Response: Thank you, we will add this reference.
L330: Dameron et al. 2024; J Micropalaeontol.
Response: Thank you, we will add this reference.
L342: Bombard et al. 2024 for the Miocene of the Ross Sea, and Seidenstein et al 2024 for the Plio-Pleistocene; both in J. Micropaleontol.
Response: Thank you, we will add this references.
L363: This is a really important observation: That planktic forams can be drawn deep under ice shelves; not likely living, but their tests can be transported great distances under the ice shelves. Dameron et al. 2024 report very rare N. pachyderma in both the upper unit (Holocene) and lower unit (late Miocene in age) of RISP; the latter helping to constrain the likely late Miocene age of the till. Likewise, Leckie and Webb 1986 and Bombard et al. 2024 report multiple occurrences of very rare planktic forams in Miocene age core samples at DSDP Site 270 and IODP Site U1521.
Response: Thank you, we will highlight this observation by adding references to modern observations from Amery Ice Shelf (Hemer et al. 2007) and fossil records cited in your comment.
L371: I'm not so sure. Bart et al., Prothero et al. and Majewski et al. have all clearly documented the presence of Globocassidulina subglobosa (or descendent G. biora) as the dominant mineralized benthic foram at and near GZWs in both the western and eastern Ross Sea. Bombard et al. and Seidenstein et al. have hypothesized that G. subglobosa is a proxy for warm CDW (or mCDW), which is why we think it's associated with the retreating GZWs. Both these studies have demonstrated the association of G. subglobosa-dominated benthic foram assemblages with incursions of warmer water planktic forams into the Ross Sea during the Miocene and Plio-Pleistocene, respectively. It would be interesting to see this type of DNA analysis conducted on the RISP cores, both the upper and lower unit (the latter is a till and likely barren of DNA but worth a try).
Response: In our statement, we referred to our sedaDNA results. This will be made clearer.
L414: I agree; why aren't these other taxa represented unless monothalamids so greatly dominate?
Response: We discuss the reasons down from line 427, for Rotaliida and Textulariida specifically in lines 439-446.
L425: This G. scitula result does not surprise me at all based on the numerous species of subpolar, temperate, and even subtropical (G.. ruber!) we have found in our Ross Sea core samples. I reported G. trilobus in the basal Miocene at DSDP Site 270, as well as other warmer water species (Leckie and Webb 1986), many warmer water species documented in the Miocene of U1521 (Bombard et al 2024), and G. ruber and G. inflata in the upper Pliocene of U1523, among other subpolar species until 1.82 Ma (Seidenstein et al. 2024). Although we have documented many different planktic species, I have not yet seen scitula., but your results do not surprise me.
Response: You are right. Scitula’s DNA was noted in open-marine facies, when only a few calcareous tests survive. We sentence will be corrected to be more neutral: “but a second planktonic species (Globorotalia scitula) is also noted in the SH results, which presence was not identified in the fossil record."
L461: Please briefly explain why ultra-short markers are more accurate for paleoecological studies.
Response: Inaccurate wording; “Thus, the newly designed ultra-short marker appears to be potentially more useful for paleoecological studies.”
Fig 2: Cores KC03, KC04, and KC18 bear striking resemblance to the upper unit of the cores recovered at RISP (Ross Ice Shelf Project of the late 1970s); specifically, agglutinated benthics dominate the top of the cores (green unit here) with very rare calcareous benthic specimens and very few planktic Neogloboquadrina pachyderma, overlying an unconformity with sparse calcareous benthic foram and diatom assemblages of mixed Miocene ages in the lower unit at RISP (brown unit here) (please see Harwood et al. 1989, Mar. Micropaleontology; Dameron et al. 2024, J. Micropalaeontology). However, the ages of the till here (brown unit) are clearly not Miocene. Have DNA analyses of the RISP cores been investigated?
Response: The dominance of agglutinated foraminifera in upper open-marine sections of cores from deeper (> ca. 500 m) parts of the Ross Sea is a common feature (Majewski et al. 2018, 2020), regardless of the type of underlying sediments. It is simply a consequence of elevated CCD. DNA studies should be conducted on freshly collected material. Samples are preferred to be collected directly after coring or from frozen cores. For this reason, RISP cores are rather not suited for such analyses, if we understood the last sentence of this comment correctly.
Any diatom biostratigraphy or diatom assemblages from the lower (brown and blue) units of these cores?
Response: No, no diatom data.
Fig. 3: Please define what the ST and SH datasets are in this figure caption.
Response: Will be done.
Fig. 4: Please define what the ST and SH datasets are in this figure caption.
Response: As above.
Thank you once more!
Citation: https://doi.org/10.5194/egusphere-2024-2824-AC3
-
AC3: 'Reply on RC2', Wojciech Majewski, 18 Dec 2024
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 204 | 69 | 67 | 340 | 50 | 7 | 7 |
- HTML: 204
- PDF: 69
- XML: 67
- Total: 340
- Supplement: 50
- BibTeX: 7
- EndNote: 7
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1