the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 19 Jun 2023
Aquaplanet simulations with winter and summer hemispheres: The setup and circulation response to warming
Abstract. To support further understanding of circulation changes in a warming climate, an idealised aquaplanet model setup containing summer and winter hemispheres is presented and the results of circulation changes under warming are discussed. First, a setup is introduced that allows aquaplanet simulations with a warmer and a colder hemisphere, with realistic looking summer and winter jet streams, storm tracks and precipitation patterns that are similar as in observations, with a more intense and more equatorward storm track in the winter compared to the summer hemisphere. The sea surface temperature (SST) distribution used is inspired by the June-July-August zonal mean SST found in reanalysis data, and is flexible to allow control of the occurrence of a single or double Inter-Tropical Convergence Zone (ITCZ). The setup is then used to investigate circulation changes under uniform warming, motivated by recently discussed research questions. First, we show that jet stream waviness decreases under warming when compared on isentropes with maximum wind speed or on isentropes at similar height in pressure space. Jet stream waviness increases under warming when compared at similar-valued isentropes, but solely because the corresponding isentrope is closer to the surface in the warmer climate and waviness increases downward in the atmosphere. However, we also observe a waviness increase at isentropes at very high levels (e.g., 350 K) in the colder hemisphere, which does not appear to be due to a change in height. A detailed analysis of the changes in wave amplitude for different wave numbers confirms that the amplitude of large waves increases with warming, while that of short waves decreases. The reduction in wave amplitude of short synoptic waves is found to dominate in the jet core region, where jet waviness also decreases, and is more pronounced on the equatorward side of the jet. Long waves increase in amplitude on the poleward side of the jet and at upper stratospheric levels, consistent with increased jet waviness at these levels. The projected increased amplitude of planetary waves and the reduced amplitude of synoptic waves are thus clearly apparent in our aquaplanet simulations and thus do not require zonal asymmetries or regional warming patterns. During so-called high-amplitude wave events, there is no evidence for a preferential phase of Rossby waves of wavenumbers five or seven, indicating the crucial role of stationary waves forced by orography or land-sea contrast in setting previously reported occurrences of preferential phases. Finally, we confirm that feature-based block detection requires significant tuning to the warmer climate to avoid the occurrence of spurious trends. After adjustment for changes in tropopause height, the block detection used here shows no trend in the summer hemisphere and an increase in blocking in the colder hemisphere. We also confirm previous findings that the number of surface cyclones tends to decrease globally under warming and the cyclone lifetimes become shorter, except for very long-lived cyclones.
-
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
(2718 KB)
-
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(2718 KB) - Metadata XML
- BibTeX
- EndNote
- Final revised paper
Journal article(s) based on this preprint
Interactive discussion
Status: closed
-
RC1: 'Reviewer comment on egusphere-2023-1196', Anonymous Referee #1, 19 Jul 2023
This study explores the variations in atmospheric circulation changes in winter and summer in a warming climate using an idealized aquaplanet set-up, in which the underlying prescribed SST distribution creates a warmer summer hemisphere and colder winter hemisphere. The study focuses on diagnosing changes in tropospheric wave activity, and to a lesser extent blocking and storm track metrics.
Overall, I find the authors’ aquaplanet set-up to be an intriguing and effective tool to understand the complicated dynamical changes that can occur in a warming climate. It’s great to see that the authors can reproduce many of the same changes that occur in more comprehensive GCMs in an idealized model, suggesting that future analysis of these runs could help to better disentangle the underlying mechanisms.
I see two weaknesses of the study as presently written. First, a detailed discussion of the mean circulation changes with warming seems warranted and is notably absent. Understanding how well the model captures seasonal changes in the mean circulation with warming is important before examining highly derived circulation metrics for extreme weather (waviness, blocking, etc.). Second, the writing is quite sloppy throughout and needs to be improved (see large list of typos below). I would encourage the authors to do a thorough proofreading of their analyses and text before re-submitting to eliminate any careless errors in the final manuscript.
Major Revision:
While the authors’ focus here is on understanding changes in extreme weather patterns with warming in summer and winter, it is important for readers to understand how well this aquaplanet model reproduces the mean circulation changes with warming. Some discussion of how the jet stream strength and position changes with warming (and a comparison with results from comprehensive GCMs) seems warranted. For example, if the model doesn’t produce a realistic poleward jet/storm track shift with warming, then why should we trust the results of more highly derived circulation metrics? (From Fig. 4, it appears that the jets do shift poleward with warming, but this is not discussed.) The authors focus a lot on vertical shifts in the circulation with warming, but addressing meridional shifts is also relevant.
Particularly interesting to me is whether this idealized aquaplanet experiment can reproduce the asymmetry in poleward jet shifts between winter and summer hemispheres seen in comprehensive GCMs. In particular, the Southern Hemisphere and North Atlantic jets experience much larger jet shifts under warming in summer versus winter (see Fig. 12 of Barnes and Polvani 2013). If the authors’ aquaplanet model can reproduce this result, it could help to disentangle mechanisms responsible for the seasonality of the mean circulation changes to warming as well.
Barnes, E. A., and L. Polvani, 2013: Response of the Midlatitude Jets, and of Their Variability, to Increased Greenhouse Gases in the CMIP5 Models. J. Climate, 26, 7117–7135
Minor Revisions:
Lines 30-31: I disagree with the authors’ assessment that many climate change trends, including those associated with Arctic amplification, are largest in summer. Arctic amplification is actually the smallest in summer and largest in winter (see recent review by Previdi et al. 2021, https://iopscience.iop.org/article/10.1088/1748-9326/ac1c29). Furthermore, projected blocking and storm track trends with climate change are also very pronounced in winter. The general focus on the summer hemisphere throughout the paper is not particularly well motivated in my opinion.
Line 100, and hereafter: APE – please define acronym
Line 130: “same height in pressure space” – This wording is confusing. Do you mean the same fixed vertical pressure level?
Lines 148, 152: Wavenumbers 4-8 or 5-8? Please be consistent in your methodology, or explain why there is a difference.
Line 269: I don’t see the agreement with reanalysis data here. The jets in the aquaplanet simulations look substantially stronger than in the reanalysis.
Section 4.4: To be consistent with the previous subsections in Section 4, a discussion of the blocking results from the idealized model in the context of previous results from comprehensive GCMs is notably absent here.
Lines 419-420: How do you determine whether a region is affected by a cyclone. Presumably, some radius of influence from the cyclone center is defined. It would be good to specify this in the methodology.
Line 521: From Fig. 11, it doesn’t appear that there is much difference in the decreases in cyclone lifetimes between the summer and winter hemispheres.
Figure 3: This is an odd unit for precipitation. Why not use mm/day?
Typos:
Line 29: weakened
Lines 41-42: This is not a complete sentence and does not make sense as written. Please rewrite.
Line 58: extent
Line 67: North Atlantic jet or storm track? – something is missing here
Line 77: per se
Line 80: different methods
Line 85: cause
Line 104: investigated?
Line 151: What is arctan2? Perhaps the 2 is a typo.
Line 159: 10,000
Line 165: Delete “(southern)” here, as that information is given on Line 169.
Line 187: 35,000 …. 32,000
Line 187: schemes
Line 232: double
Line 234: experiments
Lines 241-242: Not a complete sentence
Line 287: occur
Line 429: simulations
Line 447: 7,000
Line 454: West (not Central Pacific) to be consistent with Fig. 2
Line 473: held constant
Line 479: lower stratospheric (100 hPa in Fig. 6 is in lower stratosphere)
Line 483: waviness
Lines 529-530: allow us to study
Line 544, Fig. A1b caption: Text states difference is between 330 K and 340 K isentropic levels, but figure shows 340K and 350K isentropes.
Fig. 2 caption: blue contours (not blue shading)
Fig. 2 caption: flatt -> flat
Fig. 6 caption, second line: respectively
Fig. 9, legend: 30˚S for winter
Table 1: Please use commas not apostrophes in writing the numbers greater than 10,000.
Citation: https://doi.org/10.5194/egusphere-2023-1196-RC1 -
AC2: 'Reply on RC1', Sebastian Schemm, 13 Oct 2023
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-1196/egusphere-2023-1196-AC2-supplement.pdf
-
AC2: 'Reply on RC1', Sebastian Schemm, 13 Oct 2023
-
RC2: 'Comment on egusphere-2023-1196', Anonymous Referee #2, 06 Sep 2023
Schemm and Röthlisberger developed a setup for aquaplanet experiments (APE), that incorporates hemispheric asymmetries (winter and summer hemisphere). They use this setup to study a range of behaviours and phenomena, mostly related to mid-latitude jet and eddy activity, in parameter regimes corresponding to “current” and “warmer” climates, as well as single and double ITCZ settings.
This is an overall interesting study nicely showing a sample of what can be done with rather simple models. Many of the aspects that are investigated relate to highly relevant and somewhat open research questions regarding climate change and general circulation behaviour. Key results include that previously reported increase in jet-waviness on a specific isentropic level in a warmer climate could be an artefact due to a vertical shift of that level and waviness at a fixed pressure might actually decrease, as well as the observation of small-scale wave weakening and large-scale waves strengthening during global warming, consistent with other studies.
The manuscript is fairly well written, the structure is clear and the figures are good. The scientific reasoning and development of conclusions seem sound. I have a few minor comments and questions for the authors, but can otherwise recommend the paper to be published.
General remarks:
- First, I have to admit that I am no expert studying APEs. The authors give information on typical APE setups (e.g. Neale and Hoskins, 2000), but I still struggle to fully understand the advantage of the hemispheric asymmetry in the “new” simulations. I feel like most of the analyses could have been performed in two separate symmetric runs (potentially with half the runtime even, keeping the computational costs equal). Is the suggested advantage simply to allow for asymmetric double ITCZ runs? In that case, I think some additional justification of why this (tropical) asymmetry should affect mid-latitude jet and eddy behaviour might be helpful. Again, maybe this only requires some further clarification.
- I think many of the figures could benefit from row and column labels. The information is mostly given in the caption, but this could help the reader take in the content.
- Maybe I missed something, but I don’t understand why you would expect any zonal asymmetries in terms of a preferred wave phase (Sec. 4.3)? Using purely zonally symmetric boundary conditions the mean response should (given sufficient statistics) also be perfectly symmetric, right?
Specific remarks:
L100: Please introduce the acronym APE
L134 “see below”: Maybe better reference the specific section
Eq. 1: Since you are interested in the vertical structure, I was just wondering what happens if you normalise your Delta A_k with some density factor (or equivalently study changes in wave energy). Not sure if this leads to any insights or different conclusions, but might be worth considering.
L148 “we first compute”: The order doesn’t really matter here, but ok.
L152-153: Is this supposed to refer to A_k rather than Phi_hov,k? Otherwise, I don’t understand what the condition means in terms of a phase. I it is about the amplitude, you could also move this paragraph to the previous subsection.
L165: Remove “(southern)” as you mention SH blocks at the end of the paragraph
L166: Not sure if the PVU explanation is needed, but if you want to include it I would rather move it to the end of the sentence or so.
Eq. 4+5 and L228-229: You could consider to introduce separate parameters for NH and SH (like phi_0,N and phi_0,S or so). That would make the equations look a bit more complicated but might make the description of the overall setup a bit easier to understand.
Fig. 1: Could you motivate your choices of SST distribution? Later you distinguish between East and West Pacific, do you find your profiles to match these regions?
L280: The reference (Fig. 5a) seems to refer more to the previous sentence.
L290: You discuss Fig. 5 before Fig. 4, maybe swap them or change the discussion?
L338: *isentropic slope (to be more precise)
Fig. 7: What exactly does the “weighting” refer to? Also, please add sample sizes to the caption.
Citation: https://doi.org/10.5194/egusphere-2023-1196-RC2 -
AC1: 'Reply on RC2', Sebastian Schemm, 13 Oct 2023
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-1196/egusphere-2023-1196-AC1-supplement.pdf
Interactive discussion
Status: closed
-
RC1: 'Reviewer comment on egusphere-2023-1196', Anonymous Referee #1, 19 Jul 2023
This study explores the variations in atmospheric circulation changes in winter and summer in a warming climate using an idealized aquaplanet set-up, in which the underlying prescribed SST distribution creates a warmer summer hemisphere and colder winter hemisphere. The study focuses on diagnosing changes in tropospheric wave activity, and to a lesser extent blocking and storm track metrics.
Overall, I find the authors’ aquaplanet set-up to be an intriguing and effective tool to understand the complicated dynamical changes that can occur in a warming climate. It’s great to see that the authors can reproduce many of the same changes that occur in more comprehensive GCMs in an idealized model, suggesting that future analysis of these runs could help to better disentangle the underlying mechanisms.
I see two weaknesses of the study as presently written. First, a detailed discussion of the mean circulation changes with warming seems warranted and is notably absent. Understanding how well the model captures seasonal changes in the mean circulation with warming is important before examining highly derived circulation metrics for extreme weather (waviness, blocking, etc.). Second, the writing is quite sloppy throughout and needs to be improved (see large list of typos below). I would encourage the authors to do a thorough proofreading of their analyses and text before re-submitting to eliminate any careless errors in the final manuscript.
Major Revision:
While the authors’ focus here is on understanding changes in extreme weather patterns with warming in summer and winter, it is important for readers to understand how well this aquaplanet model reproduces the mean circulation changes with warming. Some discussion of how the jet stream strength and position changes with warming (and a comparison with results from comprehensive GCMs) seems warranted. For example, if the model doesn’t produce a realistic poleward jet/storm track shift with warming, then why should we trust the results of more highly derived circulation metrics? (From Fig. 4, it appears that the jets do shift poleward with warming, but this is not discussed.) The authors focus a lot on vertical shifts in the circulation with warming, but addressing meridional shifts is also relevant.
Particularly interesting to me is whether this idealized aquaplanet experiment can reproduce the asymmetry in poleward jet shifts between winter and summer hemispheres seen in comprehensive GCMs. In particular, the Southern Hemisphere and North Atlantic jets experience much larger jet shifts under warming in summer versus winter (see Fig. 12 of Barnes and Polvani 2013). If the authors’ aquaplanet model can reproduce this result, it could help to disentangle mechanisms responsible for the seasonality of the mean circulation changes to warming as well.
Barnes, E. A., and L. Polvani, 2013: Response of the Midlatitude Jets, and of Their Variability, to Increased Greenhouse Gases in the CMIP5 Models. J. Climate, 26, 7117–7135
Minor Revisions:
Lines 30-31: I disagree with the authors’ assessment that many climate change trends, including those associated with Arctic amplification, are largest in summer. Arctic amplification is actually the smallest in summer and largest in winter (see recent review by Previdi et al. 2021, https://iopscience.iop.org/article/10.1088/1748-9326/ac1c29). Furthermore, projected blocking and storm track trends with climate change are also very pronounced in winter. The general focus on the summer hemisphere throughout the paper is not particularly well motivated in my opinion.
Line 100, and hereafter: APE – please define acronym
Line 130: “same height in pressure space” – This wording is confusing. Do you mean the same fixed vertical pressure level?
Lines 148, 152: Wavenumbers 4-8 or 5-8? Please be consistent in your methodology, or explain why there is a difference.
Line 269: I don’t see the agreement with reanalysis data here. The jets in the aquaplanet simulations look substantially stronger than in the reanalysis.
Section 4.4: To be consistent with the previous subsections in Section 4, a discussion of the blocking results from the idealized model in the context of previous results from comprehensive GCMs is notably absent here.
Lines 419-420: How do you determine whether a region is affected by a cyclone. Presumably, some radius of influence from the cyclone center is defined. It would be good to specify this in the methodology.
Line 521: From Fig. 11, it doesn’t appear that there is much difference in the decreases in cyclone lifetimes between the summer and winter hemispheres.
Figure 3: This is an odd unit for precipitation. Why not use mm/day?
Typos:
Line 29: weakened
Lines 41-42: This is not a complete sentence and does not make sense as written. Please rewrite.
Line 58: extent
Line 67: North Atlantic jet or storm track? – something is missing here
Line 77: per se
Line 80: different methods
Line 85: cause
Line 104: investigated?
Line 151: What is arctan2? Perhaps the 2 is a typo.
Line 159: 10,000
Line 165: Delete “(southern)” here, as that information is given on Line 169.
Line 187: 35,000 …. 32,000
Line 187: schemes
Line 232: double
Line 234: experiments
Lines 241-242: Not a complete sentence
Line 287: occur
Line 429: simulations
Line 447: 7,000
Line 454: West (not Central Pacific) to be consistent with Fig. 2
Line 473: held constant
Line 479: lower stratospheric (100 hPa in Fig. 6 is in lower stratosphere)
Line 483: waviness
Lines 529-530: allow us to study
Line 544, Fig. A1b caption: Text states difference is between 330 K and 340 K isentropic levels, but figure shows 340K and 350K isentropes.
Fig. 2 caption: blue contours (not blue shading)
Fig. 2 caption: flatt -> flat
Fig. 6 caption, second line: respectively
Fig. 9, legend: 30˚S for winter
Table 1: Please use commas not apostrophes in writing the numbers greater than 10,000.
Citation: https://doi.org/10.5194/egusphere-2023-1196-RC1 -
AC2: 'Reply on RC1', Sebastian Schemm, 13 Oct 2023
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-1196/egusphere-2023-1196-AC2-supplement.pdf
-
AC2: 'Reply on RC1', Sebastian Schemm, 13 Oct 2023
-
RC2: 'Comment on egusphere-2023-1196', Anonymous Referee #2, 06 Sep 2023
Schemm and Röthlisberger developed a setup for aquaplanet experiments (APE), that incorporates hemispheric asymmetries (winter and summer hemisphere). They use this setup to study a range of behaviours and phenomena, mostly related to mid-latitude jet and eddy activity, in parameter regimes corresponding to “current” and “warmer” climates, as well as single and double ITCZ settings.
This is an overall interesting study nicely showing a sample of what can be done with rather simple models. Many of the aspects that are investigated relate to highly relevant and somewhat open research questions regarding climate change and general circulation behaviour. Key results include that previously reported increase in jet-waviness on a specific isentropic level in a warmer climate could be an artefact due to a vertical shift of that level and waviness at a fixed pressure might actually decrease, as well as the observation of small-scale wave weakening and large-scale waves strengthening during global warming, consistent with other studies.
The manuscript is fairly well written, the structure is clear and the figures are good. The scientific reasoning and development of conclusions seem sound. I have a few minor comments and questions for the authors, but can otherwise recommend the paper to be published.
General remarks:
- First, I have to admit that I am no expert studying APEs. The authors give information on typical APE setups (e.g. Neale and Hoskins, 2000), but I still struggle to fully understand the advantage of the hemispheric asymmetry in the “new” simulations. I feel like most of the analyses could have been performed in two separate symmetric runs (potentially with half the runtime even, keeping the computational costs equal). Is the suggested advantage simply to allow for asymmetric double ITCZ runs? In that case, I think some additional justification of why this (tropical) asymmetry should affect mid-latitude jet and eddy behaviour might be helpful. Again, maybe this only requires some further clarification.
- I think many of the figures could benefit from row and column labels. The information is mostly given in the caption, but this could help the reader take in the content.
- Maybe I missed something, but I don’t understand why you would expect any zonal asymmetries in terms of a preferred wave phase (Sec. 4.3)? Using purely zonally symmetric boundary conditions the mean response should (given sufficient statistics) also be perfectly symmetric, right?
Specific remarks:
L100: Please introduce the acronym APE
L134 “see below”: Maybe better reference the specific section
Eq. 1: Since you are interested in the vertical structure, I was just wondering what happens if you normalise your Delta A_k with some density factor (or equivalently study changes in wave energy). Not sure if this leads to any insights or different conclusions, but might be worth considering.
L148 “we first compute”: The order doesn’t really matter here, but ok.
L152-153: Is this supposed to refer to A_k rather than Phi_hov,k? Otherwise, I don’t understand what the condition means in terms of a phase. I it is about the amplitude, you could also move this paragraph to the previous subsection.
L165: Remove “(southern)” as you mention SH blocks at the end of the paragraph
L166: Not sure if the PVU explanation is needed, but if you want to include it I would rather move it to the end of the sentence or so.
Eq. 4+5 and L228-229: You could consider to introduce separate parameters for NH and SH (like phi_0,N and phi_0,S or so). That would make the equations look a bit more complicated but might make the description of the overall setup a bit easier to understand.
Fig. 1: Could you motivate your choices of SST distribution? Later you distinguish between East and West Pacific, do you find your profiles to match these regions?
L280: The reference (Fig. 5a) seems to refer more to the previous sentence.
L290: You discuss Fig. 5 before Fig. 4, maybe swap them or change the discussion?
L338: *isentropic slope (to be more precise)
Fig. 7: What exactly does the “weighting” refer to? Also, please add sample sizes to the caption.
Citation: https://doi.org/10.5194/egusphere-2023-1196-RC2 -
AC1: 'Reply on RC2', Sebastian Schemm, 13 Oct 2023
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-1196/egusphere-2023-1196-AC1-supplement.pdf
Peer review completion
Journal article(s) based on this preprint
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 312 | 108 | 17 | 437 | 8 | 9 |
- HTML: 312
- PDF: 108
- XML: 17
- Total: 437
- BibTeX: 8
- EndNote: 9
Viewed (geographical distribution)
| Country | # | Views | % |
|---|---|---|---|
| Switzerland | 1 | 122 | 28 |
| United States of America | 2 | 85 | 19 |
| Germany | 3 | 53 | 12 |
| Netherlands | 4 | 20 | 4 |
| Canada | 5 | 14 | 3 |
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
- 122
Cited
Sebastian Schemm
Matthias Röthlisberger
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(2718 KB) - Metadata XML