the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 24 Mar 2022
Pacific Decadal Oscillation modulates the Arctic sea-ice loss influence on the mid-latitude atmospheric circulation in winter
Guillaume Gastineau
Claude Frankignoul
Vladimir Lapin
Pablo Ortega
Abstract. The modulation of the winter impacts of Arctic sea ice loss by the Pacific Decadal Oscillation (PDO) is investigated in the IPSL-CM6A-LR ocean-atmosphere general circulation model. Ensembles of simulations are performed with constrained sea ice concentration corresponding to pre-industrial, present-day and future states, and initial conditions sampling warm and cold phases of the PDO. Using a general linear model, we estimate the simulated winter impact of sea ice loss, PDO and their combined effects. In response to sea ice loss, the Arctic lower troposphere warms and a negative North-Atlantic oscillation like pattern appears together with a weak deepening of the Aleutian Low. The two patterns are associated with a weakening of the poleward flank of the eddy-driven jet, while in the stratospheric the polar vortex weakens. Besides, a warm PDO phase induces a large positive Pacific North America pattern, as well as a small negative Arctic oscillation pattern associated with a weakening of the stratospheric polar vortex. However, the effects of PDO and Arctic sea ice loss are not additive. The Arctic sea ice teleconnections in both troposphere and stratosphere are reduced by the PDO, most importantly in the stratosphere. The results are discussed and compared to those obtained with the same model in atmosphere-only simulations, where sea ice loss does not significantly alter the stratospheric polar vortex.
-
Notice on discussion status
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
-
Preprint
(3351 KB)
-
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(3351 KB) - BibTeX
- EndNote
- Final revised paper
Journal article(s) based on this preprint
Amélie Simon et al.
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2022-69', Nicholas Tyrrell, 20 Apr 2022
Review for 'Pacific Decadal Oscillation modulates the Arctic sea-ice loss influence on the mid-latitude atmospheric circulation in winter', Simon et al. 2022.
Summary
This paper used coupled and atmospheric-only model experiments to investigate the interaction of Arctic sea ice loss and the PDO. A range of sensitivity experiments were performed where sea ice was artificially reduced. It was shown that the mid-latitude and stratospheric response to sea ice loss was similar to the response to the warm phase of the PDO. A linear regression algorithm was used to determine that the effect of sea ice and PDO was not additive, and the atmospheric response to sea ice loss was dampened by the PDO.
Overall, I thought this was a well written paper with interesting results. I recommend publication with minor revisions. Mostly I have only minor comments and clarifications, however, I was unsure about the results in Figure 9 which seemed to contradict earlier key results.
General comments:
Figure 9: If the combined effect of future sea ice and the PDO is not additive, and the PDO dampens the sea ice response (e.g. Figs. 3-7), why does Figure 9 not show a difference between the gradient of the lines in Fig. 9? For example, Fig. 4, bottom row, shows a difference in the combined Aleutian Low/PDO response between PD:PDO and FUT:PDO, since it's a linear regression would that also mean that the response of the PDO is dampened by reduced sea ice (apologies if I am misunderstanding this)? Which would imply that in PD should have a steeper gradient that FUT in Fig. 9.
My other general comment concerns the PDO in the experimental setup, if the experiments consist of 200 x 12 month periods, how can there be decadal variability?
Minor comments:
Line 32: North Atlantic Oscillation (NAO)
Line 34: "while in the stratosphere the polar vortex weakens"
Introduction: Is there a hypothesis for the mechanism for reduced sea ice -> negative NAO, and could that be briefly stated in the introduction?
Line 74: "late autum"
Line 227: Why were concatenated outputs used for the EOF analysis? Taking the second EOF as a physical mode can be problematic (e.g. Dommenget & Latif, 2002 https://doi.org/10.1175/1520-0442(2002)015%3C0216:ACNOTI%3E2.0.CO;2) and, as stated in the text, the first EOF is due to the reduced sea ice, so would taking the EOF from each experiment first and combining them be better?
Line 228: "This EOF analysis uses the member dimension instead of the time dimension, as classically used" Since each member is one year, I assume it's equivalent to using annual means, is that right? Could that be briefly specified.
Line 252: "... winter, defined as the 3-month mean in December-February-March" Is that meant to be Dec-Jan-Feb, or is there a reason for excluding January from the winter mean?
Line 269 and 273: "βPD is the regression coefficient determining the effect of the sea ice in PD (FUT) when compared to PI (same for βFUT with FUT);". To make it easier to read, I recommend separating the description of βPD and βFUT into two sentences rather than using parentheses.
Line 455: Define DCPP
Line 477: "Concerning the amplitude of the response to sea ice loss". The amplitude of what?
Citation: https://doi.org/10.5194/egusphere-2022-69-RC1 - AC1: 'Reply on RC1', Amelie Simon, 03 Jun 2022
-
RC2: 'Comment on egusphere-2022-69', Anonymous Referee #2, 06 May 2022
Summary & General comments: This is an interesting take on the multitude of studies concerning the atmospheric response to Arctic sea-ice loss where a general linear model is used to determine the response to sea-ice loss and PDO separately, and the interaction between them. The fact that PDO and sea-ice loss interact non-lineary is certainly of great interest. To me it was a novel technique that I believe should be introduced to the community, but consequently, I do think it could use a bit more clarification and fleshing out of both the methods and the discussion, and for that reason I have selected minor revisions, but also included a heading below with "major" just to separate out some of the more substantial changes I'd like to see from the smaller ones.
Minor comments:
Throughout:
Inconsistent use of sea-ice loss vs. sea ice loss: my preference is the former.
Adding ‘the’ in front of sea-ice loss is not usually necessary.
Abstract
Perhaps it’s worth mentioning that these are PAMIP-style experiments in the abstract?
L33: I find "Weak deepening" is sort of awkward wording, perhaps something like “small increase in strength of the Aleutian Low”, or “modest deepening…”.
L34: In the stratospheric the polar vortex -> the stratospheric polar vortex/in the stratosphere…
L35: Besides: hard to know what is meant here. Is it saying that on the other hand the PDO does X or in addition, it does X?
L38: I was confused which phase of the PDO you were referring to, upon reading the paper I understood. It could be worth mentioning that it is for both phases of the PDO here as a result of the method used.
Introduction
L85: There are quite a few more coupled studies you haven’t referenced here, e.g. McCusker et al 2017, Oudar et al 2017, Blackport & Kushner 2016,2017, Hay et al 2022, England et al 2020, Sun et al 2018….
L87: Perhaps include a reference to to Smith et al 2020 on the NAO signal to noise paradox.
L92 I believe the Cvijanovic study uses a slab ocean, so it might not be the best example to include here as Deser et al 2015 showed that the nature of the slab ocean response was quite different. Hay et al 2022 also discusses the deepening Aleutian Low response and PDO-like response of SST’s.
L108: What is meant by December Wave 1? Wave 1 pattern in december or something else?
L119: 'oppositely' as in they will cancel each other out? Where does the cancellation occur?
L125: extension -> extent
L127: Many of the responses to sea-ice loss? There have been some efforts to make multi-model comparisons (Screen et al 2018, Hay et al 2022, though I appreciate that IPSL hasn’t been used before so I understand what is meant. Though it occurs to me that since these are PAMIP experiements it would be possible to extend this analysis to other models that have done the coupled runs as well.
Methodology
L145: CM6 -> CM6A
L154: resolution increases
L175: which procedures are meant here?
L200: This last sentence probably belongs in the previous paragraph
L231 & elsewhere: FU->FUT
L241-: I’m a bit unclear on this due to the wording here, (esp. together with the equatorial SST anomaly), does this just mean that this is a known bias of climate models that they extend too far westward?
L261-262: This might be easier to read if written as two equations (assuming I’m understanding it correctly, the dummy variables effectively makes this two separate equations?)
L275: residue -> residual
L281: FDR is becoming more common within climate science but I think this bit will be a bit opaque to most readers and needs clarifying
Results
L310: Since I took issue with using 'weak deepening' in the abstract, here I want to note that you don't use it in the text, so perhaps it's not even necessary to state in that way in the abstract.
L406: What do you mean by somehow? Somewhat?
Summary & Discussion
See comment above in Intro section about other studies I think should be referenced here, and my thought son some of the discrepancies.
L446: How so?
L478: Increase in what?
L503: missing -LR
L506: Besides again, not clear what it meant.
Last paragraph: some ocean analysis is also done in Hay et al 2022.
Figures:
Fig 1: Fut-PI, and PD-PI surely? As there’s a [-] in SIC? Or does that mean unitless?
Fig 8. Perhaps changing the vertical extent/scaling in the bottom panels would be helpful for making this easier to read
Fig 9. I’m a bit confused by what is meant by the triangles being associated with terciles, it looks as though they’re just located at -1, 0, 1?
Major
L71: “is likely” might be a bit of an overstatement/oversimplification of the large body of research debating the topic and considering how small of an effect is found (e.g. Smith et al 2022), as well this just being one driver of mid-latitude climate change, where change driven by lower latitudes may be more important and induce changes of the opposite sign (e.g. tug of war between high/low latitudes)
L257: Can this be shown? For readers not familiar with the method this might be helpful. This is an example of where I think readers would appreciate a bit more of an explanation of the method,, though I understand there’s a limited amount of space. I think it’s a very interesting and, to me, novel way to analyse these types of experiments. I’m not sure what sort of assumptions go in to this, for example.
L297: So this suggests agrees with the results of Screen & Francis 2016, but shouldn't that effect have been quite hard to detect with present day sea-ice loss? I think discussing your results here in the context of theirs might be useful.
I like that there is a discussion of the short length of the experiments, as this is an atypical way of running coupled experiments and is sure to reduce oceanic effects, particularly slow time-scale ones. Perhaps it’s not surprising that the results are not too different between ATM and CPL, but it feels like something is missing between what is stated in the abstract about comparing ATM and CPL experiments as this is just a single paragraph and doesn’t really delve in to the effects of the stratospheric weakening.
References:
Smith, D.M., Scaife, A.A., Eade, R. et al. North Atlantic climate far more predictable than models imply. Nature 583, 796–800 (2020). https://doi.org/10.1038/s41586-020-2525-0
Hay, S., P. Kushner, R. Blackport, K. McCusker, T. Oudar, L. Sun, M. England, C. Deser, J. Screen and L. Polvani (2022), Separating the influences of low-latitude warming and sea ice loss on Northern Hemisphere climate change, Journal of Climate
England, M., L. Polvani, L. Sun and C. Deser (2020), Tropical climate responses to projected Arctic and Antarctic sea-ice loss, Nature Geoscience, 13, 275-281, doi: 10.1038/s41561-020-0546-9
Oudar, T., Sanchez-Gomez, E., Chauvin, F. et al. Respective roles of direct GHG radiative forcing and induced Arctic sea ice loss on the Northern Hemisphere atmospheric circulation. Clim Dyn 49, 3693–3713 (2017). https://doi.org/10.1007/s00382-017-3541-0
Sun, L., Alexander, M., & Deser, C. (2018). Evolution of the Global Coupled Climate Response to Arctic Sea Ice Loss during 1990–2090 and Its Contribution to Climate Change, Journal of Climate, 31(19), 7823-7843
Citation: https://doi.org/10.5194/egusphere-2022-69-RC2 - AC2: 'Reply on RC2', Amelie Simon, 03 Jun 2022
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2022-69', Nicholas Tyrrell, 20 Apr 2022
Review for 'Pacific Decadal Oscillation modulates the Arctic sea-ice loss influence on the mid-latitude atmospheric circulation in winter', Simon et al. 2022.
Summary
This paper used coupled and atmospheric-only model experiments to investigate the interaction of Arctic sea ice loss and the PDO. A range of sensitivity experiments were performed where sea ice was artificially reduced. It was shown that the mid-latitude and stratospheric response to sea ice loss was similar to the response to the warm phase of the PDO. A linear regression algorithm was used to determine that the effect of sea ice and PDO was not additive, and the atmospheric response to sea ice loss was dampened by the PDO.
Overall, I thought this was a well written paper with interesting results. I recommend publication with minor revisions. Mostly I have only minor comments and clarifications, however, I was unsure about the results in Figure 9 which seemed to contradict earlier key results.
General comments:
Figure 9: If the combined effect of future sea ice and the PDO is not additive, and the PDO dampens the sea ice response (e.g. Figs. 3-7), why does Figure 9 not show a difference between the gradient of the lines in Fig. 9? For example, Fig. 4, bottom row, shows a difference in the combined Aleutian Low/PDO response between PD:PDO and FUT:PDO, since it's a linear regression would that also mean that the response of the PDO is dampened by reduced sea ice (apologies if I am misunderstanding this)? Which would imply that in PD should have a steeper gradient that FUT in Fig. 9.
My other general comment concerns the PDO in the experimental setup, if the experiments consist of 200 x 12 month periods, how can there be decadal variability?
Minor comments:
Line 32: North Atlantic Oscillation (NAO)
Line 34: "while in the stratosphere the polar vortex weakens"
Introduction: Is there a hypothesis for the mechanism for reduced sea ice -> negative NAO, and could that be briefly stated in the introduction?
Line 74: "late autum"
Line 227: Why were concatenated outputs used for the EOF analysis? Taking the second EOF as a physical mode can be problematic (e.g. Dommenget & Latif, 2002 https://doi.org/10.1175/1520-0442(2002)015%3C0216:ACNOTI%3E2.0.CO;2) and, as stated in the text, the first EOF is due to the reduced sea ice, so would taking the EOF from each experiment first and combining them be better?
Line 228: "This EOF analysis uses the member dimension instead of the time dimension, as classically used" Since each member is one year, I assume it's equivalent to using annual means, is that right? Could that be briefly specified.
Line 252: "... winter, defined as the 3-month mean in December-February-March" Is that meant to be Dec-Jan-Feb, or is there a reason for excluding January from the winter mean?
Line 269 and 273: "βPD is the regression coefficient determining the effect of the sea ice in PD (FUT) when compared to PI (same for βFUT with FUT);". To make it easier to read, I recommend separating the description of βPD and βFUT into two sentences rather than using parentheses.
Line 455: Define DCPP
Line 477: "Concerning the amplitude of the response to sea ice loss". The amplitude of what?
Citation: https://doi.org/10.5194/egusphere-2022-69-RC1 - AC1: 'Reply on RC1', Amelie Simon, 03 Jun 2022
-
RC2: 'Comment on egusphere-2022-69', Anonymous Referee #2, 06 May 2022
Summary & General comments: This is an interesting take on the multitude of studies concerning the atmospheric response to Arctic sea-ice loss where a general linear model is used to determine the response to sea-ice loss and PDO separately, and the interaction between them. The fact that PDO and sea-ice loss interact non-lineary is certainly of great interest. To me it was a novel technique that I believe should be introduced to the community, but consequently, I do think it could use a bit more clarification and fleshing out of both the methods and the discussion, and for that reason I have selected minor revisions, but also included a heading below with "major" just to separate out some of the more substantial changes I'd like to see from the smaller ones.
Minor comments:
Throughout:
Inconsistent use of sea-ice loss vs. sea ice loss: my preference is the former.
Adding ‘the’ in front of sea-ice loss is not usually necessary.
Abstract
Perhaps it’s worth mentioning that these are PAMIP-style experiments in the abstract?
L33: I find "Weak deepening" is sort of awkward wording, perhaps something like “small increase in strength of the Aleutian Low”, or “modest deepening…”.
L34: In the stratospheric the polar vortex -> the stratospheric polar vortex/in the stratosphere…
L35: Besides: hard to know what is meant here. Is it saying that on the other hand the PDO does X or in addition, it does X?
L38: I was confused which phase of the PDO you were referring to, upon reading the paper I understood. It could be worth mentioning that it is for both phases of the PDO here as a result of the method used.
Introduction
L85: There are quite a few more coupled studies you haven’t referenced here, e.g. McCusker et al 2017, Oudar et al 2017, Blackport & Kushner 2016,2017, Hay et al 2022, England et al 2020, Sun et al 2018….
L87: Perhaps include a reference to to Smith et al 2020 on the NAO signal to noise paradox.
L92 I believe the Cvijanovic study uses a slab ocean, so it might not be the best example to include here as Deser et al 2015 showed that the nature of the slab ocean response was quite different. Hay et al 2022 also discusses the deepening Aleutian Low response and PDO-like response of SST’s.
L108: What is meant by December Wave 1? Wave 1 pattern in december or something else?
L119: 'oppositely' as in they will cancel each other out? Where does the cancellation occur?
L125: extension -> extent
L127: Many of the responses to sea-ice loss? There have been some efforts to make multi-model comparisons (Screen et al 2018, Hay et al 2022, though I appreciate that IPSL hasn’t been used before so I understand what is meant. Though it occurs to me that since these are PAMIP experiements it would be possible to extend this analysis to other models that have done the coupled runs as well.
Methodology
L145: CM6 -> CM6A
L154: resolution increases
L175: which procedures are meant here?
L200: This last sentence probably belongs in the previous paragraph
L231 & elsewhere: FU->FUT
L241-: I’m a bit unclear on this due to the wording here, (esp. together with the equatorial SST anomaly), does this just mean that this is a known bias of climate models that they extend too far westward?
L261-262: This might be easier to read if written as two equations (assuming I’m understanding it correctly, the dummy variables effectively makes this two separate equations?)
L275: residue -> residual
L281: FDR is becoming more common within climate science but I think this bit will be a bit opaque to most readers and needs clarifying
Results
L310: Since I took issue with using 'weak deepening' in the abstract, here I want to note that you don't use it in the text, so perhaps it's not even necessary to state in that way in the abstract.
L406: What do you mean by somehow? Somewhat?
Summary & Discussion
See comment above in Intro section about other studies I think should be referenced here, and my thought son some of the discrepancies.
L446: How so?
L478: Increase in what?
L503: missing -LR
L506: Besides again, not clear what it meant.
Last paragraph: some ocean analysis is also done in Hay et al 2022.
Figures:
Fig 1: Fut-PI, and PD-PI surely? As there’s a [-] in SIC? Or does that mean unitless?
Fig 8. Perhaps changing the vertical extent/scaling in the bottom panels would be helpful for making this easier to read
Fig 9. I’m a bit confused by what is meant by the triangles being associated with terciles, it looks as though they’re just located at -1, 0, 1?
Major
L71: “is likely” might be a bit of an overstatement/oversimplification of the large body of research debating the topic and considering how small of an effect is found (e.g. Smith et al 2022), as well this just being one driver of mid-latitude climate change, where change driven by lower latitudes may be more important and induce changes of the opposite sign (e.g. tug of war between high/low latitudes)
L257: Can this be shown? For readers not familiar with the method this might be helpful. This is an example of where I think readers would appreciate a bit more of an explanation of the method,, though I understand there’s a limited amount of space. I think it’s a very interesting and, to me, novel way to analyse these types of experiments. I’m not sure what sort of assumptions go in to this, for example.
L297: So this suggests agrees with the results of Screen & Francis 2016, but shouldn't that effect have been quite hard to detect with present day sea-ice loss? I think discussing your results here in the context of theirs might be useful.
I like that there is a discussion of the short length of the experiments, as this is an atypical way of running coupled experiments and is sure to reduce oceanic effects, particularly slow time-scale ones. Perhaps it’s not surprising that the results are not too different between ATM and CPL, but it feels like something is missing between what is stated in the abstract about comparing ATM and CPL experiments as this is just a single paragraph and doesn’t really delve in to the effects of the stratospheric weakening.
References:
Smith, D.M., Scaife, A.A., Eade, R. et al. North Atlantic climate far more predictable than models imply. Nature 583, 796–800 (2020). https://doi.org/10.1038/s41586-020-2525-0
Hay, S., P. Kushner, R. Blackport, K. McCusker, T. Oudar, L. Sun, M. England, C. Deser, J. Screen and L. Polvani (2022), Separating the influences of low-latitude warming and sea ice loss on Northern Hemisphere climate change, Journal of Climate
England, M., L. Polvani, L. Sun and C. Deser (2020), Tropical climate responses to projected Arctic and Antarctic sea-ice loss, Nature Geoscience, 13, 275-281, doi: 10.1038/s41561-020-0546-9
Oudar, T., Sanchez-Gomez, E., Chauvin, F. et al. Respective roles of direct GHG radiative forcing and induced Arctic sea ice loss on the Northern Hemisphere atmospheric circulation. Clim Dyn 49, 3693–3713 (2017). https://doi.org/10.1007/s00382-017-3541-0
Sun, L., Alexander, M., & Deser, C. (2018). Evolution of the Global Coupled Climate Response to Arctic Sea Ice Loss during 1990–2090 and Its Contribution to Climate Change, Journal of Climate, 31(19), 7823-7843
Citation: https://doi.org/10.5194/egusphere-2022-69-RC2 - AC2: 'Reply on RC2', Amelie Simon, 03 Jun 2022
Peer review completion
Journal article(s) based on this preprint
Amélie Simon et al.
Amélie Simon et al.
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 280 | 116 | 17 | 413 | 7 | 8 |
- HTML: 280
- PDF: 116
- XML: 17
- Total: 413
- BibTeX: 7
- EndNote: 8
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(3351 KB) - Metadata XML