the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Resilient Antarctic monsoonal climate prevented ice growth during the Eocene
Peter Bijl
Anna von der Heydt
Appy Sluijs
Henk Dijkstra
Abstract. Understanding the extreme greenhouse of the Eocene (56–34Ma ago) is key to anticipate potential future conditions. During the Eocene, the Antarctic continent remained mostly ice-free despite large temperature swings. Seemingly contradictory indications of ice and thriving vegetation complicate modelling efforts to explain the Antarctic Eocene climate. We use global climate model simulations to show that extreme seasonality mostly limited ice growth. Without ice sheets, much of the Antarctic continent saw monsoonal conditions. Perennially mild and wet conditions along Antarctic coastlines support vegetation reconstructions, while extreme seasonality elsewhere promoted intense weathering shown in proxy records. The results can thus explain the coexistence of warm and wet conditions in some regions, while small ice caps could form near the coast. The resilience of the climate regimes seen in these simulations agrees with the longevity of warm Antarctic conditions during the Eocene, but also challenge our view on glacial inception.
- Preprint
(6687 KB) - Metadata XML
-
Supplement
(13150 KB) - BibTeX
- EndNote
Michiel Baatsen et al.
Status: final response (author comments only)
-
RC1: 'Comment on cp-2023-36', Anonymous Referee #1, 12 Jul 2023
Overview
This paper focuses on Antarctic climate during the extreme hothouse world of the Eocene (defined here as 56-34 million years ago), with the aim of better anticipating future climate changes. A state-of-the-art global climate model is used to reproduce Eocene climate under various scenarios of CO2, and the paper concludes that extreme seasonality in Antarctic climate limits ice growth, instead allowing most of the continent to have monsoonal conditions. The results also show the resilience of the various Antarctic climate regimes.
The paper is, for the most part, well written, with a clear structure including an excellent abstract (nicely summarising the main findings), followed by an introduction, brief methodology, results and discussion section. Rather than just considering basic climatic variables, the authors use three indices - a glacial index, an evergreen vegetation index and a monsoonal index - to assess their simulations, thereby bringing a novel approach to this area of research. Combined with the state-of-the-art GCM simulations, this paper is therefore both novel and timely. My recommendation, therefore, is to accept this manuscript, subject to some minor revisions as outlined below.
Major comments
My main comment is that currently a section on model-data comparison is completely missing and is needed. Model-data comparison is mentioned briefly at the very end of the introduction, but is a single sentence only. It is mentioned again in slightly more detail at the end, in Section 4.2, but this focuses more on vegetation reconstructions. Before acceptance, I would like to see another section, perhaps immediately before the model results are presented, giving a more in-depth analysis of how key variables (e.g. temperature and precipitation) from the various simulations match the available proxy data. At the moment this is missing, which makes it harder to determine the accuracy/reliability of these simulations and therefore the conclusions.
Another, less major, comment would be that some restructuring is needed. For example, Section 3.5 appears to overlap with Section 4.4; I would I would recommend moving 3.5 down into the discussion and merging with 4.4. Likewise Section 3.4 appears to discuss the same matter as Section 4.3, and therefore could be merged into the latter discussion section. Likewise Section 4.1 appears to be a summary of the main findings, which appears to overlap with Section 4.6 and again could be merged into one section at the end.
Minor comments
- Abstract: I would like to see a little more motivation - the authors say that understanding Antarctic climate during the Eocene is "key to anticipate potential future conditions”, but why?
- Results: As discussed above, this needs a new section containing a model-data comparison.
- Throughout the manuscript: brackets (or some sort of other punctuation) are needed around references e.g. from the first paragraph, it currently reads “… during the middle and late Eocene Pross et al. (2012); Contreras et al. (2013, 2014); Passchier et al. (2017); Bijl et al. (2021)” whereas something like “… during the middle and late Eocene (Pross et al. 2012; Contreras et al. 2013, 2014; Passchier et al. 2017; Bijl et al. 2021)” would be better.
- Figures: I find some of these quite hard to read/interpret, either because they use very bright and alternating, fairly garish colours (e.g. Figure 2) or they are quite small (e.g. Figure 6). Could these be re-drawn to aid interpretation?
Citation: https://doi.org/10.5194/cp-2023-36-RC1 -
AC1: 'Reply on RC1', Michiel Baatsen, 25 Sep 2023
Overview
This paper focuses on Antarctic climate during the extreme hothouse world of the Eocene (defined here as 56-34 million years ago), with the aim of better anticipating future climate changes. A state-of-the-art global climate model is used to reproduce Eocene climate under various scenarios of CO2, and the paper concludes that extreme seasonality in Antarctic climate limits ice growth, instead allowing most of the continent to have monsoonal conditions. The results also show the resilience of the various Antarctic climate regimes.
The paper is, for the most part, well written, with a clear structure including an excellent abstract (nicely summarising the main findings), followed by an introduction, brief methodology, results and discussion section. Rather than just considering basic climatic variables, the authors use three indices - a glacial index, an evergreen vegetation index and a monsoonal index - to assess their simulations, thereby bringing a novel approach to this area of research. Combined with the state-of-the-art GCM simulations, this paper is therefore both novel and timely. My recommendation, therefore, is to accept this manuscript, subject to some minor revisions as outlined below.
AR: We thank the reviewer for their constructive remarks and positive evaluation of our work. Please find our point-specific responses below.
Major comments
My main comment is that currently a section on model-data comparison is completely missing and is needed. Model-data comparison is mentioned briefly at the very end of the introduction, but is a single sentence only. It is mentioned again in slightly more detail at the end, in Section 4.2, but this focuses more on vegetation reconstructions. Before acceptance, I would like to see another section, perhaps immediately before the model results are presented, giving a more in-depth analysis of how key variables (e.g. temperature and precipitation) from the various simulations match the available proxy data. At the moment this is missing, which makes it harder to determine the accuracy/reliability of these simulations and therefore the conclusions.
AR: Our original approach was to focus results on the qualitative assessment of the simulated Eocene climate on Antarctica and refer to earlier work for proxy references.. However, we agree to the above comment and propose to add a quantitative model-proxy comparison to the results section, focussed on the Antarctic Eocene climate.
Another, less major, comment would be that some restructuring is needed. For example, Section 3.5 appears to overlap with Section 4.4; I would recommend moving 3.5 down into the discussion and merging with 4.4. Likewise Section 3.4 appears to discuss the same matter as Section 4.3, and therefore could be merged into the latter discussion section. Likewise Section 4.1 appears to be a summary of the main findings, which appears to overlap with Section 4.6 and again could be merged into one section at the end.
AR: We will restructure to avoid overlap between these sections. Adding a results section on the model-proxy comparison will likely also help to achieve this.
Minor comments
- Abstract: I would like to see a little more motivation - the authors say that understanding Antarctic climate during the Eocene is "key to anticipate potential future conditions”, but why?
AR: Fully ice-free conditions on Antarctica are of course far away from any expected near-future scenario, but are still indicative of what can be expected in terms of high latitude warmth in a more general sense. We will add some more motivation/nuance on this part.
- Results: As discussed above, this needs a new section containing a model-data comparison.
AR: we agree that this can help to support the model results.
- Throughout the manuscript: brackets (or some sort of other punctuation) are needed around references e.g. from the first paragraph, it currently reads “… during the middle and late Eocene Pross et al. (2012); Contreras et al. (2013, 2014); Passchier et al. (2017); Bijl et al. (2021)” whereas something like “… during the middle and late Eocene (Pross et al. 2012; Contreras et al. 2013, 2014; Passchier et al. 2017; Bijl et al. 2021)” would be better.
AR: We will fix this.
- Figures: I find some of these quite hard to read/interpret, either because they use very bright and alternating, fairly garish colours (e.g. Figure 2) or they are quite small (e.g. Figure 6). Could these be re-drawn to aid interpretation?
AR: Thank you for pointing this out; we will reconsider colouring and optimise readability.
Citation: https://doi.org/10.5194/cp-2023-36-AC1
-
RC2: 'Comment on cp-2023-36', Anonymous Referee #2, 28 Aug 2023
The manuscript authored by Baatsen et al. delves into a comprehensive exploration of the climate conditions prevalent in ice-free Antarctica during the Late Eocene and Early Oligocene epochs. The paper accentuates the pivotal role played by seasonality in shaping the climate of this ancient Antarctic landscape. While the prevailing temperature and precipitation patterns appear conducive to the proliferation of temperate or subtropical flora along the coastal regions, aligning well with empirical observations, an intriguing contrast emerges when examining the conditions further inland. Inland areas seem to present a paradoxical scenario: too cold during the winters to permit the growth of vegetation and yet too warm during the summers to facilitate glacial inception, unless we consider the presence of formidable high mountain ranges. The paper is notably commendable for its clarity and well-crafted prose. Its structure is clear.
However, despite its strengths, I do have a few comments and suggestions to offer."
The formatting of bibliographic references in the manuscript appears to be incorrect and requires correction.
L25: “a different representation of coastal waters”: please clarify this sentence.
L46: A sentence explaining how climate equilibrium is defined is necessary.
L69 : I'm uncertain about how interpolation can reliably yield accurate results for the five missing Eocene cases. Additionally, there is some ambiguity in Table 1 regarding the notation 'E2+E2-E4.' It's not clear to me what this notation signifies in the context of the integration of simulations or the number of years simulated from the starting point corresponding to the end of a simulation."
L80: To my knowledge, SMB is not used in Goldner et al 2014.
L87: A reference for the evergreen vegetation index is needed
L93 : Different types of monsoon indices are available, and it's worth noting that this paper's monsoonal index, tested with ERA5 dataset, demonstrates relatively robust performance, particularly at lower latitudes. However, the results show an index close to 1 in regions such as eastern Siberia, the Pacific coast in South America, and North America. This prompts the question of whether this criterion is adequately restrictive.
Figure 2 could provide valuable insights into the wind direction, which should undergo a reversal or significant change with seasonal fluctuations in sea level pressure. Yet, understanding changes in atmospheric circulation is somewhat challenging when examining sea level pressure due to the lack of labels on isocontours, especially for the summer at 38 Ma, where thick contours are absent. To enhance clarity and insight, it is advisable to include a plot of low-tropospheric wind and sea level pressure in the Supplementary Information section."
L101 : How was the lapse rate fixed ?
Figure 4 : Add labels on isocontours.
Figure 5: in caption, “and” is missing between potential melt and precipitation
L176 and L245: The authors conclude that the prevailing temperature and precipitation conditions are conducive to the growth of temperate forests along the coast but not in the inner continent. It's worth considering whether the vegetation prescribed in the simulations (Baatsen et al. 2020) as a cool/warm mixed forest remains consistent despite the observed seasonality. Another question arises regarding the representation of the inner Antarctica continent. Given the simulations, it appears that a bare soil with a higher albedo might provide a more accurate representation of this region. It can be useful to explore the potential impact of such a change in representation on the overall conclusions of the study.
L254: The model-data comparison is very brief. This section needs to be completed.
L269: a “s” is missing in “consistent”
Citation: https://doi.org/10.5194/cp-2023-36-RC2 -
AC2: 'Reply on RC2', Michiel Baatsen, 25 Sep 2023
The manuscript authored by Baatsen et al. delves into a comprehensive exploration of the climate conditions prevalent in ice-free Antarctica during the Late Eocene and Early Oligocene epochs. The paper accentuates the pivotal role played by seasonality in shaping the climate of this ancient Antarctic landscape. While the prevailing temperature and precipitation patterns appear conducive to the proliferation of temperate or subtropical flora along the coastal regions, aligning well with empirical observations, an intriguing contrast emerges when examining the conditions further inland. Inland areas seem to present a paradoxical scenario: too cold during the winters to permit the growth of vegetation and yet too warm during the summers to facilitate glacial inception, unless we consider the presence of formidable high mountain ranges. The paper is notably commendable for its clarity and well-crafted prose. Its structure is clear.
AR: We thank the reviewer for their positive words and helpful comments. Please find our point-specific responses below..
However, despite its strengths, I do have a few comments and suggestions to offer."
The formatting of bibliographic references in the manuscript appears to be incorrect and requires correction.
AR: This will be corrected.
L25: “a different representation of coastal waters”: please clarify this sentence.
AR: This refers to the large differences in near-coastal currents and temperatures compared to most previous simulations, we will adjust to clarify.
L46: A sentence explaining how climate equilibrium is defined is necessary.
AR: we will add a sentence on the temperature trends seen in the simulations, referring the reader to earlier work for further details.
L69 : I'm uncertain about how interpolation can reliably yield accurate results for the five missing Eocene cases. Additionally, there is some ambiguity in Table 1 regarding the notation 'E2+E2-E4.' It's not clear to me what this notation signifies in the context of the integration of simulations or the number of years simulated from the starting point corresponding to the end of a simulation."
AR: Different lines of information are given in this table. First, we show simulation length for the cases that are simulated, including information on how/when they were branched off. Second, we indicate how results are being interpolated from the available simulations. Clearly, this was not presented in a clear way and so we will clarify this.
Regarding the interpolation: we indeed make a rather strong assumption of full linearity here, which is explained in the methods. Our simple methodology for this specific use is supported well by the large overall consistency between cases shown in the results. Importantly, we do not use any of these extrapolated datasets to make a detailed assessment of the Antarctic climate, and this will be clarified in the revised text.
L80: To my knowledge, SMB is not used in Goldner et al 2014.
AR: We will check and change if necessary.
L87: A reference for the evergreen vegetation index is needed
AR: The evergreen index is devised in this paper. We will add a figure to the supplementary material showing how the different climate indices capture the pre-industrial conditions globally.
L93 : Different types of monsoon indices are available, and it's worth noting that this paper's monsoonal index, tested with ERA5 dataset, demonstrates relatively robust performance, particularly at lower latitudes. However, the results show an index close to 1 in regions such as eastern Siberia, the Pacific coast in South America, and North America. This prompts the question of whether this criterion is adequately restrictive.
AR: In their current form, the indices used are indeed quite simple and therefore not completely restrictive. As indicated above, we will add a figure in the supplement to show this for pre-industrial conditions, along with motivation and discussion. Importantly, however, we do not claim that the Antarctic Eocene climate is strictly identical to that of present-day sub-tropical monsoons, because of course there are fundamental differences originating, e.g., from latitude. But there are important similarities. In fact, the present-day climate of Eastern Siberia and parts of North America may represent rather close analogues to much of Eocene Antarctica, with cold and dry winters alternating with warm and wet summers. We will add this to the discussion.
Figure 2 could provide valuable insights into the wind direction, which should undergo a reversal or significant change with seasonal fluctuations in sea level pressure. Yet, understanding changes in atmospheric circulation is somewhat challenging when examining sea level pressure due to the lack of labels on isocontours, especially for the summer at 38 Ma, where thick contours are absent. To enhance clarity and insight, it is advisable to include a plot of low-tropospheric wind and sea level pressure in the Supplementary Information section."
AR: This is indeed tricky, as MSLP gradients become rather weak overall in the Eocene summer climate on Antarctica. We agree that it would be helpful to add a visible indication of low-tropospheric flow here and will include this in the figure. As MSLP contours are repeated in the panels showing precipitation, we will replace these by quivers (some testing shows that this works well).
L101 : How was the lapse rate fixed ?
AR: The constant lapse rate of 8C/km is commonly used in the modelling of high-latitude ice sheets. This suits our overall qualitative approach and is motivated by simplicity. We will add explanation and/or referencing here.
Figure 4 : Add labels on isocontours.
AR: We will add labels to the thick contours, to improve readability.
Figure 5: in caption, “and” is missing between potential melt and precipitation
AR: We will correct this.
L176 and L245: The authors conclude that the prevailing temperature and precipitation conditions are conducive to the growth of temperate forests along the coast but not in the inner continent. It's worth considering whether the vegetation prescribed in the simulations (Baatsen et al. 2020) as a cool/warm mixed forest remains consistent despite the observed seasonality. Another question arises regarding the representation of the inner Antarctica continent. Given the simulations, it appears that a bare soil with a higher albedo might provide a more accurate representation of this region. It can be useful to explore the potential impact of such a change in representation on the overall conclusions of the study.
AR: Our results indeed put some question marks on the vegetation that was implied in the simulations. Some testing with alternative (tundra-like) vegetation types suggested only minimal impacts, with mainly a slight increase in seasonality over the continental interior. We chose not to include these in the results but will add this to the discussion.
L254: The model-data comparison is very brief. This section needs to be completed.
AR: We agree and will add a more quantitative section on model-proxy comparison to the results section.
L269: a “s” is missing in “consistent”
AR: We will correct this.
Citation: https://doi.org/10.5194/cp-2023-36-AC2
-
AC2: 'Reply on RC2', Michiel Baatsen, 25 Sep 2023
Michiel Baatsen et al.
Michiel Baatsen et al.
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 272 | 72 | 9 | 353 | 24 | 4 | 2 |
- HTML: 272
- PDF: 72
- XML: 9
- Total: 353
- Supplement: 24
- BibTeX: 4
- EndNote: 2
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1