The role of atmospheric CO<sub>2</sub> in controlling patterns of sea surface temperature change during the Pliocene

Burton, Lauren E.; Haywood, Alan M.; Tindall, Julia C.; Dolan, Aisling M.; Hill, Daniel J.; McClymont, Erin L.; Ho, Sze Ling; Ford, Heather L.

doi:https://doi.org/10.5194/cp-2023-98

Preprints

https://doi.org/10.5194/cp-2023-98

Preprints

30 Nov 2023

| 30 Nov 2023

Status: a revised version of this preprint was accepted for the journal CP and is expected to appear here in due course.

The role of atmospheric CO₂ in controlling patterns of sea surface temperature change during the Pliocene

Lauren E. Burton, Alan M. Haywood, Julia C. Tindall, Aisling M. Dolan, Daniel J. Hill, Erin L. McClymont, Sze Ling Ho, and Heather L. Ford

Abstract. We present the role of CO₂ forcing in controlling patterns of Late Pliocene sea surface temperature (SST) using seven models from Phase 2 of the Pliocene Model Intercomparison Project (PlioMIP2) and palaeoclimate proxy data from the PlioVAR working group. At a global scale, SST change in the Late Pliocene relative to the pre-industrial is predominantly driven by CO₂ forcing in the low and mid-latitudes and non-CO₂ forcing in the high latitudes. We find that CO₂is the dominant driver of SST change at the vast majority of proxy data sites assessed (17 out of 19), but the relative dominance of this forcing varies between all proxy sites, with CO₂ forcing accounting for between 27 % and 82 % of the total change seen. The dearth of proxy data sites in the high latitudes means that only two sites assessed here are predominantly forced by non-CO₂ forcing (such as changes to ice sheets and orography), both of which are in the North Atlantic Ocean.

We extend the analysis to show the seasonal patterns of SST change and its drivers at a global scale and at a site-specific level for three chosen proxy data sites. We also present a new estimate of Late Pliocene climate sensitivity using site-specific proxy data values. This is the first assessment of site-specific drivers of SST change in the Late Pliocene and highlights the strengths of using palaeoclimate proxy data alongside model outputs to further develop our understanding of the Late Pliocene. We use the best-available proxy and model data, but the sample sizes remain limited and the confidence in our results would be improved with greater data availability.

Received: 23 Nov 2023 – Discussion started: 30 Nov 2023

Download & links

Preprint (PDF, 2116 KB)

Supplement (3163 KB)

Download & links

Lauren E. Burton, Alan M. Haywood, Julia C. Tindall, Aisling M. Dolan, Daniel J. Hill, Erin L. McClymont, Sze Ling Ho, and Heather L. Ford

Status: closed

CC1:
'Comment on cp-2023-98', Tim Herbert, 11 Jan 2024

I like this approach.

There is additional alkenone data from ODP Site 1208 (W central Pacific) that you can add:
Venti, N. L., K. Billups, and T.D. Herbert, 2013, Increased sensitivity of the Plio-Pleistocene northwest Pacific to obliquity forcing, Earth and Planetary Science Letters 384: 121-131

Also, I wondered about the large anomaly in the North Atlantic. All except super-high resolution models do not get the location of the North Atlantic Current correctly in the modern day. Can this also affect your results for the comparison to 400 ppm CO2?

Tim Herbert.

Citation: https://doi.org/10.5194/cp-2023-98-CC1
- AC1: 'Reply on CC1', Lauren Burton, 12 Mar 2024
  
  Many thanks for your comment and feedback, Tim.
  For reproducibility we decided to only include sites in the published KM5c alkenone database in McClymont et al. (2020), though future research could apply this method to any site for which there is the necessary data.
  On the North Atlantic, we do see variation between the models used in this study in terms of the FCO₂ on SST as well as the Pliocene minus pre-industrial SST anomaly. It is not yet possible for us to run the resolution of model required to accurately capture the North Atlantic Current to equilibrium, so the results we have are the best approximation available given current capability. This will undoubtedly have an impact on the results, but by considering an ensemble of models (rather than an individual model) we are more aware of the differences and subsequent uncertainty. The FCO₂ method itself may also begin to explain some of the uncertainty, e.g., anomalously strong non-CO₂ forcing coinciding with anomalously strong warming (see Supplementary of Burton et al. (2023) for FCO₂ on SST in the six models considered in this paper).
  
  Citation: https://doi.org/10.5194/cp-2023-98-AC1
RC1:
'Comment on cp-2023-98', Anonymous Referee #1, 25 Jan 2024

This manuscript is a welcome addition to the Pliocene paleoclimate literature by assessing site-specific drivers of SST change. As the authors indicate, the methodology highlights the strengths of using palaeoclimate proxy data in combination with model output to better understand the mid Piacenzian and future climate. The FCO2 methodology is novel and allows a better estimate of climate sensitivity.
The manuscript is well-written, for the most part properly referenced and the methodology is clearly explained. I have only minor suggestions for the authors to consider:
Line 25: If you are going to call the mPWP an analog provide a qualifier or brief explanation- analog means different things to different readers. Also, these are good citations to recent work, but the concept of the Pliocene as an analog began at least as far back as Budyko (1982) or Zubakov and Borzenkova (1988). There is quite a literature since then about the mPWP and its relevance to future climate.
Line 34: Haywood et al. in press should be 2024.
Line 55: The last time the top of the Mammoth subchron was dated 3.220 Ma was Cande and Kent (1995). Most of us would use 3.207 Ma. I think this might be part of the PlioVAR procedure but without explanation here it just looks like you have the wrong age. You could avoid this by cutting the sentence starting on line 53: "Geological proxy data...McClymont et al. (2020)" as it doesn't really add anything to this paper.
Line 84: You are absolutely correct in the example of using isotope tracers rather than calculating temperature from the oxygen isotopes. However, since you are only considering SST and the Pliocene, is there a different example you could use? No one is estimating Pliocene SST using oxygen isotopes. You could, but you would need more explanation here.
Line 183: I think the PRISM3 time interval would be Dowsett et al. 2010, not 2016.
Other than these minor suggestions/comments, I like this paper and hope to see the methodology used as PlioMIP3 simulations become available.

Citation: https://doi.org/10.5194/cp-2023-98-RC1
- AC2: 'Reply on RC1', Lauren Burton, 12 Mar 2024
  
  The authors thank Anonymous Referee #1 for their comments and feedback. Revisions are described below and indicated in the revised manuscript.
  Additional detail and references to Budyko (1982) and Zubakov and Borzenkova (1988) have been added in Line 26/27. Full consideration of the analogue concept is beyond the scope of this paper, but work is in preparation that addresses this important point.
  The citation in Line 35, and associated entry in the reference list, has been updated from Haywood et al. (in press) to Haywood et al. (2024).
  The suggested sentence starting on Line 54 has been removed.
  The isotope example has been reworded for improved clarity in Line 90/91.
  The citation in Line 183/184 has been corrected to Dowsett et al. (2010).
  
  Citation: https://doi.org/10.5194/cp-2023-98-AC2
RC2:
'Comment on cp-2023-98', Anonymous Referee #2, 29 Feb 2024

Burton et al. applied the Fco2 method to KM5c, a crucial time period of the Pliocene targeted by the data and modeling community for intercomparison purposes. They aimed to assess the extent to which observed sea surface temperature (SST) patterns are influenced by CO2 and other forcings. The study also investigated the impact of seasonality on reconstructed mean annual changes. Additionally, site-specific calculations of equilibrium climate sensitivity (ECS) were conducted to obtain a range of Pliocene ECS values based on site location, proxy used, and the magnitude of SST changes in response to CO2 forcing.
Overall, this study presents a robust analysis using state-of-the-art data and modeling products. The manuscript is well-written, and its eventual publication in Climate of the Past is recommended. However, there are some overarching questions that the authors should address before acceptance.
The mention of "patterns of SST" in the title and throughout the manuscript raises questions about the SST pattern effect, a topic of recent interest. The authors should clarify whether the observed Pliocene warming pattern solely reflects polar amplification or is also influenced by zonal gradients or phenomena like ENSO. Recent advancements in understanding the warming patterns of other geological periods, such as the LGM, late Miocene, and PETM, could provide valuable context.
Furthermore, the manuscript may appear overly technical to readers unfamiliar with Pliocene Model Intercomparison Project (PlioMIP) terminology and concepts. It would be beneficial for the authors to articulate the fundamental questions regarding Pliocene climate change and explain how their study contributes to addressing them. The attempt to correlate data-model offsets with Fco2 may not yield significant insights, as these two factors are not inherently linked.
In addressing these points, the authors are encouraged to consider the broader implications of their findings for understanding Pliocene paleoclimate. Specifically, they should highlight how their study contributes to addressing key questions in the field and elucidating the significance of SST patterns in the Pliocene context.
Specific comments:
L85-100: While the discussions on proxy data and model simulations are pertinent, they are not directly related to the topics presented here. The authors should elaborate more on the significance of SST patterns in the broader Pliocene climate context.

Citation: https://doi.org/10.5194/cp-2023-98-RC2
- AC3: 'Reply on RC2', Lauren Burton, 12 Mar 2024
  
  The authors thank Anonymous Referee #2 for their comments and feedback. Revisions are described below and indicated in the revised manuscript.
  Additional detail has been added to Section 1 which discusses the pattern of SST change in the Pliocene, and some fundamental questions of Pliocene climate change.
  The title of the paper and references to “patterns” of SSTs have been rephrased to avoid confusion with the SST pattern effect, which is not centrally examined in this paper. “Patterns” was originally used to describe the global-scale picture of SSTs. We have now included comments on some zonal and meridional gradients seen in the Pliocene, but the focus of the paper remains on site-specific SSTs and the overall global picture. Future research could use the FCO₂ analysis to consider the drivers of zonal and meridional changes in more detail.
  Lines 85-100 detail examples of previous data-model comparison efforts and hence provide important context for the novel data-model comparison analysis presented in this paper.
  
  Citation: https://doi.org/10.5194/cp-2023-98-AC3
EC1: 'Editor Comment on cp-2023-98', Christo Buizert, 01 Mar 2024

Dear authors,
Your manuscript has now been evaluated by two reviewers. As you can see from their reports, they are both supportive of publication after you address their concerns and comments. Please provide a detailed response to the referee reports, and to the comment by Tim Herbert. I will most likely be inviting you to submit a revised manuscript, so feel free to write your response in the form of proposed changes to the manuscript.
Please let me know if you have any questions, and I look forward to reading your responses.
Best regards, Christo Buizert (CP editor)

Citation: https://doi.org/10.5194/cp-2023-98-EC1

Status: closed

CC1:
'Comment on cp-2023-98', Tim Herbert, 11 Jan 2024

I like this approach.

There is additional alkenone data from ODP Site 1208 (W central Pacific) that you can add:
Venti, N. L., K. Billups, and T.D. Herbert, 2013, Increased sensitivity of the Plio-Pleistocene northwest Pacific to obliquity forcing, Earth and Planetary Science Letters 384: 121-131

Also, I wondered about the large anomaly in the North Atlantic. All except super-high resolution models do not get the location of the North Atlantic Current correctly in the modern day. Can this also affect your results for the comparison to 400 ppm CO2?

Tim Herbert.

Citation: https://doi.org/10.5194/cp-2023-98-CC1
- AC1: 'Reply on CC1', Lauren Burton, 12 Mar 2024
  
  Many thanks for your comment and feedback, Tim.
  For reproducibility we decided to only include sites in the published KM5c alkenone database in McClymont et al. (2020), though future research could apply this method to any site for which there is the necessary data.
  On the North Atlantic, we do see variation between the models used in this study in terms of the FCO₂ on SST as well as the Pliocene minus pre-industrial SST anomaly. It is not yet possible for us to run the resolution of model required to accurately capture the North Atlantic Current to equilibrium, so the results we have are the best approximation available given current capability. This will undoubtedly have an impact on the results, but by considering an ensemble of models (rather than an individual model) we are more aware of the differences and subsequent uncertainty. The FCO₂ method itself may also begin to explain some of the uncertainty, e.g., anomalously strong non-CO₂ forcing coinciding with anomalously strong warming (see Supplementary of Burton et al. (2023) for FCO₂ on SST in the six models considered in this paper).
  
  Citation: https://doi.org/10.5194/cp-2023-98-AC1
RC1:
'Comment on cp-2023-98', Anonymous Referee #1, 25 Jan 2024

This manuscript is a welcome addition to the Pliocene paleoclimate literature by assessing site-specific drivers of SST change. As the authors indicate, the methodology highlights the strengths of using palaeoclimate proxy data in combination with model output to better understand the mid Piacenzian and future climate. The FCO2 methodology is novel and allows a better estimate of climate sensitivity.
The manuscript is well-written, for the most part properly referenced and the methodology is clearly explained. I have only minor suggestions for the authors to consider:
Line 25: If you are going to call the mPWP an analog provide a qualifier or brief explanation- analog means different things to different readers. Also, these are good citations to recent work, but the concept of the Pliocene as an analog began at least as far back as Budyko (1982) or Zubakov and Borzenkova (1988). There is quite a literature since then about the mPWP and its relevance to future climate.
Line 34: Haywood et al. in press should be 2024.
Line 55: The last time the top of the Mammoth subchron was dated 3.220 Ma was Cande and Kent (1995). Most of us would use 3.207 Ma. I think this might be part of the PlioVAR procedure but without explanation here it just looks like you have the wrong age. You could avoid this by cutting the sentence starting on line 53: "Geological proxy data...McClymont et al. (2020)" as it doesn't really add anything to this paper.
Line 84: You are absolutely correct in the example of using isotope tracers rather than calculating temperature from the oxygen isotopes. However, since you are only considering SST and the Pliocene, is there a different example you could use? No one is estimating Pliocene SST using oxygen isotopes. You could, but you would need more explanation here.
Line 183: I think the PRISM3 time interval would be Dowsett et al. 2010, not 2016.
Other than these minor suggestions/comments, I like this paper and hope to see the methodology used as PlioMIP3 simulations become available.

Citation: https://doi.org/10.5194/cp-2023-98-RC1
- AC2: 'Reply on RC1', Lauren Burton, 12 Mar 2024
  
  The authors thank Anonymous Referee #1 for their comments and feedback. Revisions are described below and indicated in the revised manuscript.
  Additional detail and references to Budyko (1982) and Zubakov and Borzenkova (1988) have been added in Line 26/27. Full consideration of the analogue concept is beyond the scope of this paper, but work is in preparation that addresses this important point.
  The citation in Line 35, and associated entry in the reference list, has been updated from Haywood et al. (in press) to Haywood et al. (2024).
  The suggested sentence starting on Line 54 has been removed.
  The isotope example has been reworded for improved clarity in Line 90/91.
  The citation in Line 183/184 has been corrected to Dowsett et al. (2010).
  
  Citation: https://doi.org/10.5194/cp-2023-98-AC2
RC2:
'Comment on cp-2023-98', Anonymous Referee #2, 29 Feb 2024

Burton et al. applied the Fco2 method to KM5c, a crucial time period of the Pliocene targeted by the data and modeling community for intercomparison purposes. They aimed to assess the extent to which observed sea surface temperature (SST) patterns are influenced by CO2 and other forcings. The study also investigated the impact of seasonality on reconstructed mean annual changes. Additionally, site-specific calculations of equilibrium climate sensitivity (ECS) were conducted to obtain a range of Pliocene ECS values based on site location, proxy used, and the magnitude of SST changes in response to CO2 forcing.
Overall, this study presents a robust analysis using state-of-the-art data and modeling products. The manuscript is well-written, and its eventual publication in Climate of the Past is recommended. However, there are some overarching questions that the authors should address before acceptance.
The mention of "patterns of SST" in the title and throughout the manuscript raises questions about the SST pattern effect, a topic of recent interest. The authors should clarify whether the observed Pliocene warming pattern solely reflects polar amplification or is also influenced by zonal gradients or phenomena like ENSO. Recent advancements in understanding the warming patterns of other geological periods, such as the LGM, late Miocene, and PETM, could provide valuable context.
Furthermore, the manuscript may appear overly technical to readers unfamiliar with Pliocene Model Intercomparison Project (PlioMIP) terminology and concepts. It would be beneficial for the authors to articulate the fundamental questions regarding Pliocene climate change and explain how their study contributes to addressing them. The attempt to correlate data-model offsets with Fco2 may not yield significant insights, as these two factors are not inherently linked.
In addressing these points, the authors are encouraged to consider the broader implications of their findings for understanding Pliocene paleoclimate. Specifically, they should highlight how their study contributes to addressing key questions in the field and elucidating the significance of SST patterns in the Pliocene context.
Specific comments:
L85-100: While the discussions on proxy data and model simulations are pertinent, they are not directly related to the topics presented here. The authors should elaborate more on the significance of SST patterns in the broader Pliocene climate context.

Citation: https://doi.org/10.5194/cp-2023-98-RC2
- AC3: 'Reply on RC2', Lauren Burton, 12 Mar 2024
  
  The authors thank Anonymous Referee #2 for their comments and feedback. Revisions are described below and indicated in the revised manuscript.
  Additional detail has been added to Section 1 which discusses the pattern of SST change in the Pliocene, and some fundamental questions of Pliocene climate change.
  The title of the paper and references to “patterns” of SSTs have been rephrased to avoid confusion with the SST pattern effect, which is not centrally examined in this paper. “Patterns” was originally used to describe the global-scale picture of SSTs. We have now included comments on some zonal and meridional gradients seen in the Pliocene, but the focus of the paper remains on site-specific SSTs and the overall global picture. Future research could use the FCO₂ analysis to consider the drivers of zonal and meridional changes in more detail.
  Lines 85-100 detail examples of previous data-model comparison efforts and hence provide important context for the novel data-model comparison analysis presented in this paper.
  
  Citation: https://doi.org/10.5194/cp-2023-98-AC3
EC1: 'Editor Comment on cp-2023-98', Christo Buizert, 01 Mar 2024

Dear authors,
Your manuscript has now been evaluated by two reviewers. As you can see from their reports, they are both supportive of publication after you address their concerns and comments. Please provide a detailed response to the referee reports, and to the comment by Tim Herbert. I will most likely be inviting you to submit a revised manuscript, so feel free to write your response in the form of proposed changes to the manuscript.
Please let me know if you have any questions, and I look forward to reading your responses.
Best regards, Christo Buizert (CP editor)

Citation: https://doi.org/10.5194/cp-2023-98-EC1

Lauren E. Burton, Alan M. Haywood, Julia C. Tindall, Aisling M. Dolan, Daniel J. Hill, Erin L. McClymont, Sze Ling Ho, and Heather L. Ford

Supplement

https://doi.org/10.5194/cp-2023-98-supplement

Lauren E. Burton, Alan M. Haywood, Julia C. Tindall, Aisling M. Dolan, Daniel J. Hill, Erin L. McClymont, Sze Ling Ho, and Heather L. Ford

Viewed

Total article views: 489 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	Supplement	BibTeX	EndNote
360	106	23	489	26	15	18

HTML: 360
PDF: 106
XML: 23
Total: 489
Supplement: 26
BibTeX: 15
EndNote: 18

Views and downloads (calculated since 30 Nov 2023)

Month	HTML	PDF	XML	Total
Dec 2023	128	40	2	170
Jan 2024	96	19	5	120
Feb 2024	33	11	2	46
Mar 2024	78	26	12	116
Apr 2024	25	10	2	37

Cumulative views and downloads (calculated since 30 Nov 2023)

Month	HTML	PDF	XML	Total
Dec 2023	128	40	2	170
Jan 2024	96	19	5	120
Feb 2024	33	11	2	46
Mar 2024	78	26	12	116
Apr 2024	25	10	2	37

Viewed (geographical distribution)

Total article views: 493 (including HTML, PDF, and XML) Thereof 493 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 26 Apr 2024

Short summary

The Pliocene (~3 million years ago) is of interest because its warm climate is similar to projections of the future. We explore the role of atmospheric carbon dioxide in forcing patterns of sea surface temperature during the Pliocene by combining climate model outputs with palaeoclimate proxy data. We investigate whether this role changes seasonally, and also use our data to suggest a new estimate of Pliocene climate sensitivity. More data are needed to further explore the results presented.


Total:	0
HTML:	0
PDF:	0
XML:	0

The role of atmospheric CO2 in controlling patterns of sea surface temperature change during the Pliocene

Supplement

Viewed

Viewed (geographical distribution)

The role of atmospheric CO₂ in controlling patterns of sea surface temperature change during the Pliocene