the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 11 Mar 2024
North American Pleistocene Glacial Erosion and Thin Pliocene Regolith Thickness Inferred from Data-Constrained Fully Coupled Ice-Climate-Sediment modelling
Abstract. Beyond the impact of glacial isostatic adjustment, landscape evolution is typically neglected at large scale when considering the basal boundary condition for ice sheet and climate modelling over past glacial cycles. Erosion and changing sediment loads impact bed elevation, land/sea mask, and basal drag. To date, how the above affects past ice sheet evolution is unclear.
Constraining the role that Pliocene regolith may have played in Pleistocene glacial cycle variability requires bounds on the amount of regolith preceding those glacial cycles. However, quantitative bounds on regolith thickness at the spatial scale of the glaciated regions of North America are currently absent in the literature.
To address the above, we present an updated sediment production and transport model with dynamically calculated soft sediment mask, isostatic adjustment to dynamical sediment load and bedrock erosion, and a new subglacial hydrology model coupled to the Glacial Systems Model. The coupled model is capable of multi-million year integrations driven only by greenhouse gas concentration and insolation. The model passes a set of verification tests and conserves mass. We assess parametric sensitivity of sediment transport rates.
We compare the final sediment solutions in an ensemble of whole-Pleistocene simulations for a range of initial (Pliocene) regolith thicknesses against multiple estimates for present day drift thickness distribution, Quaternary sediment volume in the Atlantic Ocean, and erosion depth estimates. Consistency of modelling and data constraints requires a Pliocene regolith thickness of less than 50 m.
- Preprint
(17271 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2024-620', Neil Ross, 05 Jul 2024
This is an interesting manuscript, that attempts to deploy an innovative approach to evaluate the question of how pre-glacial sediments may have influenced the flow and evolution of the early-Quaternary Laurentide Ice Sheet, and estimates the likely thickness of the pre-glacial regolith across North America.
The methods used (i.e. numerical ice sheet modelling) appear to be appropriate for a study of this nature, though I am not an ice-sheet modeller so cannot provide substantive insight regarding the appropriateness of the technicalities of the model used etc. However, I have some concerns about (a) reproducibility of the study (i.e. by others who may want to replicate the experiments); and (b) the current quality of the results and discussion sections of the manuscript. Both points will need addressing or refuting by the authors.
General comments:
- Figures, I recommend that the authors improve the manuscript figures throughout. They should certainly replace the rainbow colour schemes used in some figures
- Definitions and terms. In lines 46-49 the authors make a case that defining terminology is important, but at no point in the manuscript do they define ‘regolith’.
- Structure – methodology really begins in section2, rather than section 5. Suggest that sections 2-5 are re-titled to reflect this. Also, the start of section 2 has some sequencing issues (e.g. sediment module within the GSM is introduce before the GSM). Authors need to review their manuscript carefully to ensure that structure is optimised and logical.
- What forces the model? It was a bit unclear to me, as a non-modeller, from sections 2-5 what was actually forcing the model from a climate perspective. I assume it is the CO2 record derived in section 4, but that isn’t really stated anywhere as far as I could make out. There’s lots of technical info in sections 2-5, but I’m far from convinced that it is presented very accessibly to the majority palaeo-facing audience of Climate of the Past. Recommend that the authors review sections 2-5 carefully and look to restructure these sections and make them more accessible to the general audience of this journal.
- Reproducibility – As a non-modeller I found sections 2-5 somewhat challenging to decipher. Now, that may simply be my lack of knowledge and experience in the modelling sphere. However, even acknowledging this, the nature of sections 2-5 made me concerned as to whether another modelling team could pick up the methodology outlined in 2-5 and repeat the experiments as performed in this manuscript. I therefore suggest that the authors either (a) ensure that reproducibility of the simulations is possible by improving sections 2-5; or (b) refute my concerns in response to this review. Please be assured that I’m not attempting to be provocative here. I’m aware of my limitations regarding numerical modelling and simply want to ensure good practice and scientific reproducibility, as well as making the manuscript the best it can possibly be.
- Overall, the results section lacks a detailed quantified description of the simulations. The authors need to describe the patterns, trends and anomalies of their simulations before they compare against existing dataset and interpret what their simulations mean. At present, a description of their results and simulations is absent from the manuscript.
- The discussion section needs to be separated from the conclusions, and the ‘basic description of results’ presented in this section for the first time needs to be included in the results section. The authors need to substantively rewrite results, discussion and conclusions sections.
Specific comments:
Abstract – Lots on background (lines 1-7), and methods (lines 8-15), but findings are limited to a simple statement that regolith was 50 m thick (lines 15-16). More reporting of results, and the wider implications of these are necessary in the abstract. Look at the examples in lines 36-41 of the introduction section and explain how you have addressed such ‘wider implications’. [having now read the full manuscript, the abstract may simply be a reflection of issues further on].
Introduction – first paragraph of introduction section is totally devoid of relevant references. Please insert appropriate references to support the statements made.
Lines 44-45: probably a little too much to explain that you have a results, a discussion and a conclusion section.
Lines 46-47. Erosion does not = transport.
Lines 47-48: “quarrying and abrasion” – I presume that the authors are referring to quarrying and abrasion by glaciers here. This isn’t, to my understanding, the process that produces regolith by physical weathering. Recommend that authors re-examine lines 46-49 carefully to ensure that the meaning is correct and consistent throughout.
Line 53: I’m not sure about the preference of the journal for English or American English, but suggest change to “palaeo”.
Lines 52-55: may need some additional explanation for non-modellers, who will make up the majority of the audience for Climate of the Past.
Line 69-70: this sentence requires a reference to support the point being made?
Line 70-71: Avoid sentences like “table 1 shows…” as this is essentially what the figure caption should say. Recommend changing lines 70-72 to make the statement required, ending the sentence with “….(Table 1).”.
Line 78: ensure that tense used is consistent throughout the manuscript (e.g. it should be past tense here “Pollard and DeConto (2003) used…”)
Lines 73-97: useful to explain previous approaches to the application of numerical modelling to the problem posed, but this section would benefit from being less list-like, and more theme-driven (e.g. what is good or bad about the previous modelling work). It would also be useful to reduce the amount of technical information in these sentences, unless it represents a critical step in model development useful for the current research. It might help to reframe the sentences in this section so that they do not begin with the references first (e.g. say “Improvements and developments in model performance were made by increasing XYZ (Reference)” rather than “Reference (2020) did XYX….”). It would also be useful to restructure these sentences so that distinction is made between models applied to the specific (North American-limited) research question being addressed by this paper, and models applied elsewhere in the world (e.g. Antarctica), but which have implication for the North American research question being addressed in this manuscript.
Section 2:
Lines 104-112: carries on where 73-97 left off and is review material rather than methodology. There seems to be a blending of background/context into methods across sections 1-5, and I suggest that the start of section 2 should be clearly ‘methodology’. These lines list the specifics of the previous models, without actually critically evaluating them. Recommend that the authors state clearly what is good or bad about these previous model setups, as it isn’t clear to the interested reader at the present. I recommend that line 104 (and section 2) starts with “The model we use in this study is…..”. Perhaps also cite both Melanson 2012 and Melanson et al. 2013?
Lines 113-120: these convince me that section 2 needs to be “methodology” as the authors are explaining how they have modified an existing model for use in this study.
Section 2.1: there are sequencing issues here, as lines 104-120 talk about the GSM, but it is first detailed here. I recommend that the section on the sediment model is moved below the introduction to the GSM that it is part of (i.e. it likely sits best between current 2.1 and 2.2.
Line 146-47: again, I recommend that the authors please avoid sentences that begin like “In fig D1……”. This sentence needs reworded, and perhaps greater information provided to extend the point being made.
Figure 1: I recommend that the authors consider replacing the rainbow colour scheme used in fig 1a. Please see https://theconversation.com/how-rainbow-colour-maps-can-distort-data-and-be-misleading-167159 for why.
Line 149: the “basal velocity” of what? Please be more specific here and look for similar issues elsewhere in section 2.
Section 2.5: The title of this section is very strange “Sediment Transport Sensitivity for a Square Root Ice Sheet”. Are you sure this is the most appropriate heading? Why not just drop the heading for section 2.5 and merge it with 2.4 “Sediment Model Verification Tests”?
Figure 2 – the colour legend labels (i.e. melt m/a) are overprinted.
Section 3
Lines 209-240: Again, a bit list-y and lacks a certain degree of critical evaluation. Suspect this section could be condensed and be re-focused around the key messages (e.g. maps are consistent in some places, some areas lack data, there are challenges as to whether superficial sediments = Quaternary, or Quaternary+, there are issues with different methodologies and datasets – e.g. seismic data vs boreholes etc.) rather than Reference 1 said X, Reference 2 said Y.
Figure 3: recommend that colour scheme is replaced with another more accessible colour scheme.
Lines 241-254: much less list-like. Recommend 209-240 are rewritten to be more like this.
Section 4:
Heading: I recommend that this be changed to better reflect the content of the section, which is about more than just “uncertainty”.
Line 267: this section needs to state at the outset that the model is driven by the atmospheric CO2 curve the authors derive in this section. It should also explain a bit more about how it is deployed in the modelling. There’s lots of detail in this section (and possibly sections 2-3?) about how the CO2 forcing is derived, but there is a shortfall as to how it is used.
Section 5:
Finally, we arrive at a section titled “methodology”, though in reality the methodology section started in section 2. I would argue that section 5 should really be titled “model tuning and simulation runs”. I also encourage the authors to ensure that the full methods section (i.e. current sections 2-5) can ensure reproducibility if others wanted to repeat the simulations (see ‘general comments’ above).
Section 6 (Results):
Line 318: It is not terribly good practice to begin a results section stating that you compare the model results with something else. I recommend that the authors describe the model simulations before comparing them to other datasets. The audience needs to be led through the results before comparing against other existing datasets. Describe the patterns, trends and anomalies in the model simulations first before anything else.
Lines 318-325: Reads like methods text rather than a description of results.
Lines 318-384: this section needs rewritten to actually describe the model simulation output, and needs to be far less dependent on published literature.
Figure 7: this is the first figure of the results section. It seems to be a very high-level type of figure (i.e. comparing model derived regolith values with present-day sediment thicknesses) to be the first figure of results. My reading of this figure, and lines 318-384 is that several ‘interim’ model simulation outputs, and their description, are missing. It would be useful to see some model output before this meta-data style analysis.
Section 6.1.1: I am very confused about the heading numbering for the results section. There is no 6.1.2. Why not just make 6.1.1, 6.2 and then make current 6.2 as 6.3? Section 6.1.2. is dominated by ‘methods’ material. Only the final short paragraph (lines 424-427) is truly results description.
Section 6.2: A fragment (a single paragraph), that does not effectively describe figure 9.
Section 7 (Discussion and conclusions):
I strongly recommend that the authors carefully review sections 6&7. There is material in section 7 that is ‘basic’ description of data and the figures and should be in section 6. There is some content in section 7 that is well argued and important, but it needs to be separated from the description of results and improved. It also needs to be grounded on much stronger foundations (i.e. a robust results section) than it currently is now.
The manuscript requires a separate conclusions section.
Professor Neil Ross
Newcastle University
5th July 2024
Citation: https://doi.org/10.5194/egusphere-2024-620-RC1 -
RC2: 'Reply on RC1', Neil Ross, 05 Jul 2024
Additional note - the manuscript includes many references to geographic locations. None of these (even Hudson Bay) are labelled on a figure. The authors should show where these locations are, otherwise the interested reader cannot follow the arguments made (e.g. in discussion section).
Citation: https://doi.org/10.5194/egusphere-2024-620-RC2
-
RC3: 'Comment on egusphere-2024-620', Anonymous Referee #2, 19 Jul 2024
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2024/egusphere-2024-620/egusphere-2024-620-RC3-supplement.pdf
-
RC4: 'Comment on egusphere-2024-620', Anonymous Referee #3, 16 Aug 2024
This article integrates landscape evolution characteristics, such as erosion and sediment transport, to evaluate the evolution of ice sheet characteristics over the Pleistocene. This is a highly relevant advancement in establishing the response of landscape, glaciers, and sea-level rise to climate variations.
Despite the potential relevance and novelty of the work, I found the manuscript poorly organized and the methodology unclear, making the work not reproducible and difficult to understand. Additionally, there seem to be issues with the LaTex compilation so numerous equations and references are not given- I do not know why the manuscript was submitted or sent for review with these issues. This leads me to have deeper concerns about many other aspects of the work as well.
I believe that the manuscript is not publishable in its current state and presenting this research successfully requires a very different organization and rewriting.
As stated in another comment regarding the manuscript, there seem to be two papers here. One is the model presentation even some notable findings about the model behavior presented (i.e. Fig 2 and sensitivity tests can say a lot about the behavior of the coupled system). The second is the model application and comparison with observed regolith thickness- which I believe needs some more work to link with the larger landscape/glacier/GIA system than is presented in this version.
While there are many matters that proof-reading would address, below I list several examples that are the basis for my assessment and could be useful to the authors:
General Comments:
-It is unclear to me what experiments the authors conducted or even how the model parameterized certain processes. Sections 2-5 are overlapping and should be reorganized. Examples include:
Section 2- I would start with presenting the GSM. It is very unclear to me what the basis is for the sediment model, i.e. how is sediment production and transport quantified? is it Melanson (2012) or Hildes (2021)? Better yet, the processes and their relationship could be put in the text.
Section 2.4 - 2.5 I believe this is some kind of "numerical experiment design" section. Should be streamlined combined? Also, there are some results in this section. While interesting and relevant, they belong in a separate section of the paper.
-It seems that much (important) information about the model in the appendix. I believe that this should probably be in the main text (see comment about the two papers above).
-While many experiments were conducted, it is difficult for me to understand what information or larger findings should be gleaned. For instance, is there a relationship between regolith thickness and GIA that emerges?
-As regolith thickness, along with ice dynamics and GIA are modeled together (I think), I am curious why the interaction between these processes is not discussed. Or, what is missed if regolith thickness is not evaluated?
Particular comments:Section 1.1- I think that this section should be shortened and streamlined into the information necessary to establish the basis for the modeling study here.
Ln 98-102- This introductory paragraph describes the goals of the model development, but does not establish the application of the model that is conducted later in the paper.
Ln 139- What is PD?
Equation 2: What is "L"? and how is it different from "IA"?
Section 3. Is this somehow "validation data"? If so, it should be integrated into an experiment description section. There are also numerous unsupported and uncited statements in the text that may not be relevant to the research objectives here. From reading Ln 318- it seems that only Taylor et al. 2023 was used.
Figure 1: What is this figure doing here? This is some kind of data input, that is separate from the model presentation.
Ln 301: What is a "stronger obliquity response"?
Ln 326- Couldn't model poor model performance presented here come from the difficulty in representing the complex processes with a cumbersome model and a lot of parameters? I would be more curious about how the different processes interact generally.
Figure 9: I did not find "Total Weathering" clearly defined. Is it the combination of quarrying and abrasion?
Section 6. The first paragraph of this section belongs in the "Experiment Design" section, or something like that.
Section 6.2- This examines a "snapshot" of model behavior, but it is unclear which one. For instance, does quarrying/abrasion behavior vary with the parameters tested?
Ln 450- more citations to support this section would be useful.
Figure 10: Somehow the "top" result. Why was this not presented and discussed earlier?
Citation: https://doi.org/10.5194/egusphere-2024-620-RC4
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 501 | 154 | 32 | 687 | 30 | 23 |
- HTML: 501
- PDF: 154
- XML: 32
- Total: 687
- BibTeX: 30
- EndNote: 23
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1