the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 30 May 2025
Relating extratropical atmospheric heat transport to cyclone life cycle characteristics and numbers in Southern Hemispheric winter
Abstract. Extratropical cyclones, which develop and propagate in regions known as storm tracks, account for the majority of poleward atmospheric heat transport (AHT) outside the tropics. This allows the intensity and position of the zonal mean storm track to be constrained by the hemispheric-wide radiative budget. Yet, the zonal mean nature of this approach masks the contribution of individual cyclones, which can locally constitute extreme eddy AHT. In this study, we adopt a cyclone-centered perspective to quantify in detail the relationship between poleward AHT and the life cycle characteristics of extratropical cyclones in Southern Hemispheric winter. Specifically, we combine objectively identified surface cyclone tracks derived from ERA5 data (1981–2021) with a moist static energy (MSE) framework that features an eddy-mean decomposition of the circulation. The eddy MSE flux maximizes during the cyclone intensification phase and is largest in the warm sector with a secondary maximum in the cold sector. Importantly, a considerable fraction of the flux in the warm-sector is located well equatorward of the cyclone center and thus outside the cyclonic region identified by the tracking algorithm. This leads to a latitudinal shift between maxima in cyclone frequency and eddy MSE fluxes. To bridge the gap between zonally integrated AHT and contributions from individual cyclones, local vertically integrated eddy MSE flux events are attributed to cyclones based on spatial overlap criteria. When integrating zonally and over the cyclone lifetime, the most intense cyclones (and not the ones intensifying most rapidly) generally exhibit the largest cyclone-attributed eddy MSE fluxes. Despite a disproportionate contribution to the cyclone-attributed eddy MSE fluxes by the most intense cyclones on the seasonal scale, the relationship between the seasonal number of intense cyclones and the poleward eddy MSE flux is sensitive to the choice of the eddy-mean decomposition. This result indicates that low wavenumber background flows mask the influence of cyclone characteristics in the vertical, zonal, and seasonal integral. Notably however, the relationship between the overall cyclone number and total AHT shows a peak at 50° S, which resembles the dominance of synoptic waves at this latitude in terms of AHT while the relevance of planetary waves increases poleward. Further research on the interplay between synoptic and planetary MSE fluxes in the vicinity of cyclones is needed to understand to which extent the cyclone number, which is projected to decrease under warming, could be constrained by the global energy budget.
Competing interests: At least one of the (co-)authors is a member of the editorial board of Weather and Climate Dynamics.
Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.- Preprint
(19163 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2025-2314', Anonymous Referee #1, 17 Jul 2025
The authors investigate the relationship between extratropical cyclone properties and atmospheric heat transport in the Southern Hemisphere winter. To study this relationship, the authors utilize many different methodologies, including cyclone tracking, cyclone-centered composites, and quantifying the atmospheric heat transport. The authors find that a large fraction of cyclone-related eddy atmospheric heat transport is done by intense cyclones. The authors also find that the relationship between cyclone number and atmospheric heat transport is sensitive to the definition of the variables.
While I think this manuscript attempts to answer very interesting questions as proposed in the introduction, from the current version of the manuscript, it is difficult to evaluate whether those are sufficiently addressed. The amount of analyses included in the paper, including the various methods used here, is beyond what is typically presented in a single paper. Hence, delivering the key messages to the readers while not making them lost in the technical details and discussions is important. In the current version of the manuscript, there are many places that needs to be shortened or removed, even with five appendix sections. I am familiar with the methodological details and the Southern Hemisphere storm tracks, but it was not easy going through the manuscript. Considering that the paper will be read by a broader audience, I recommend that the authors majorly restructure the manuscript. Moreover, I have some comments on the interpretation of the scientific results.
1. There are details in the results section that are not new (introduction materials) and involve in-depth discussions (discussion materials). These are helpful, but they obscure the key messages the authors are trying to deliver. In the current manuscript, all the details of the figures are explained. I appreciate the author's thoroughness, but some details can sometimes be neglected, for the sake of a bigger key message. Below, I will list parts of the result section that had particularly dense details to read.
1.1. Lines 253-260: The results are largely consistent with theoretical models of baroclinic instability. These descriptions can be shorter.
1.2. Lines 260-301: Please shorten these paragraphs so that the key message becomes clearer.
1.3 Lines 311-315: It seems that the underlying physics is the Clasius-Clapeyron equation. I also think these could be shorter.
1.4. Lines 316-323: The precipitation pattern is generally consistent with Lq flux. It was confusing to see the moisture budget brought in to explain the simple result.
1.5. Lines 353-370: The attribution methods were explained in the method section. It was clear why the authors chose to use the `nearby flux' method over `cyclone mask' method. If not comparing the two methods is not the main objective of this study, I suggest moving these (and Fig. 6) to the appendix.2. One of the major results emphasized in section 5 is that strongly-intensifying cyclones account for a disproportionate amount of cyclone-related eddy AHT. I'm not sure if this is surprising as the authors describe it. The reasoning in the manuscript is that the strongly-intensifying cyclones account for 43%, similar to the moderate cyclones' contribution, although the former is half in number. But I'm not sure if the expectation here (i.e., a proportionate result) for the strongly-intensifying cyclones is to contribute 25%. This would mean that there is no relation between cyclone intensity (or intensification rate) and AHT. If one assumes that AHT and intensity (or intensification rate) are linearly correlated (a simple y=x relationship), the top 25% of cyclones would contribute 44% to AHT, very similar to the result in the paper. Since the number is expected from a linear relationship, from some perspective, this could just be a proportionate contribution. I would recommend that authors reassess their point. Additionally, the definition of strongly-intensifying or strong cyclones change from section 3 to section 5, please provide more justification for this change.
3. Figure 11. Please provide the p-values of the linear correlation. If authors are using simple linear regression, R^2=rho^2, and additional information will be helpful to understand the significance of the relations.
4. Figure 8. I suggest that Fig. 8 be turned into a bar chart or something similar. The text does not describe latitudinal structure, and it is hard to distinguish the lines.Citation: https://doi.org/10.5194/egusphere-2025-2314-RC1 -
RC2: 'Comment on egusphere-2025-2314', Anonymous Referee #2, 01 Aug 2025
Summary
This manuscript tries to evaluate and characterize the contribution of extratropical cyclones to the total poleward atmospheric heat transport, focusing on the Southern Hemisphere winter. This is done using ERA5 reanalysis data and identifying cyclones based on local minima of sea level pressure associated with eddies. The amount of poleward heat transport associated with the cyclone is evaluated using an attribution method, based on the overlap in longitude and latitude between the surface cyclone and the lower tropospheric poleward heat transport. The main findings are: (1) Poleward heat transport by cyclones maximizes during the intensification period. (2) Poleward heat transport is much stronger on the eastward side (warm sector) of the cyclone, compared to the westward side (cold sector). (3) Latent heat transport makes a higher portion of the total transport by the cyclone as the latitude decreases and exhibits a strong asymmetry between the warm and cold sectors. (4) Intense cyclones and strongly intensifying cyclones transport much more heat poleward during their life cycle compared to weak cyclones. (5) The total poleward eddy heat flux is not correlated with the number of intense cyclones in the interannual variability, when eddies are defined as deviations from the monthly mean, yet there is a positive correlation when eddies are defined using a high-pass filter.
Each of the results is examined carefully, while considering the sensitivity to the method of identifying eddies, to the attribution method and the percentile threshold chosen for that, to the choices of latitudes and pressure levels, to the focus on intense or intensifying cyclones and to the eddy lifetime.
The manuscript is well-written and the presentation is clear and pleasant to look at, while serving the message well. The motivation for each analysis presented is explained and the connection between the results and conclusions is clear.
I therefore recommend on publication of this manuscript in Weather and Climate dynamics. I think it would be of interest to the community studying Earth’s energy budget and the role of extratropical cyclones in that budget.
I have a few minor comments elaborated below, relating mostly to a few sentences where I found it difficult to understand the intention. Also, I think it may be possible to shorten the manuscript a bit, without losing the important messages.
Comments
- Line 49: MSE includes also potential energy.
- Lines 123-124: “tracks that are… weaker than 990 hPa… are discarded” – do you mean that the SLP is higher than 990 hPa?
- Line 161: delete “for” (“for to explore”).
- Line 211: delete “4 and 5” (maybe this was supposed to refer to a certain section?).
- Lines 216, 220 and figure 2: The use of different terms to describe the stages of the life cycle is confusing: mature stage (line 216), min. SLP (figure 2), intensification phase and deepening phase (line 220). It is not clear if the mature stage and min. SLP are the same and what is the difference between the intensification and deepening stages.
- Figure 3: The discussion of the figure in lines 240-252 mentions regions of negative MSE flux. I see there are blue regions in the figure, but it is difficult to identify them exactly and connect them to the description in the text. I suggest to add a zero contour to the figure, to highlight these regions.
- Discussion of figure 8a,b in lines 381-385: The figure shows that sorting the cyclones by intensity makes the greatest difference for the MSE flux poleward of 50S, yet the authors say that this differentiator is equally important to the intensification. They argue that the high ratio between MSE flux by the largest and lowest intensity cyclones in high latitudes is due to SLP climatologically decreasing towards the pole. I don’t how that explains this ratio. The MSE flux is by transient eddies, meaning that the climatology is subtracted from the total fields. What am I missing?
- It would help to mention how the MSE flux in figure 8 is related to the MSE flux in figure 6. To me it was confusing that the total MSE flux by transient eddies in figure 6 is almost 2 PW, whereas in figure 8c it is around 0.1 PW. Is that because the results in figure 8 should be multiplied by the number of cyclones to get the result in figure 6? What is the average number of cyclones per season?
- Discussion of figure 8c,d in lines 386-394: there is no discussion of the differentiating by intensity (dotted lines).
- Lines 395-396: It is not clear to me why using this subset makes the comparison fairer (relating to comment 7).
- Lines 413-414: I find it difficult to understand the arguments in this sentence. It is not straightforward to see the connection to the correlation between poleward propagation of eddies and MSE flux.
- Lines 502-503: I find it difficult to understand the explanation for the low correlation in figure 11d and the comparison between figures 11c and f.
- Line 505: should be Fig.E1b-h instead of Fig.E1b,h.
- General comment to section 5: I appreciate the authors effort in examining the sensitivity of the results to various methods and parameters. However, this makes the paper a bit too long and heavy. I think it would not reduce the quality and completeness of the paper if some of the content is removed. To me figure D2 was useful, but other figures in the appendices (D1, D3-D6, E2) and some of the discussion in section 5 were too much loaded with details, which really make it difficult to follow. In general, I think that the effects of the method (defining eddies as deviation from the monthly mean, using a high-pass filter or considering zonal anomalies) and the choice of the percentile on the results can be explained qualitatively with less details. If the authors still prefer to keep the section as it is, that’s also fine.
- Lines 619-620: I don’t understand the argument in this sentence.
- Lines 641-643: I don’t understand the argument in this sentence.
Citation: https://doi.org/10.5194/egusphere-2025-2314-RC2 -
RC3: 'Comment on egusphere-2025-2314', Anonymous Referee #3, 12 Aug 2025
This manuscript aims to assess the link between extratropical cyclones and moist static energy (MSE) transport in the Southern Hemisphere during winter. Using ERA5 reanalysis and methods that partition energy fluxes into transient and stationary components, the authors attribute deviations from the mean to cyclones and quantify their contribution to total eddy MSE flux. The study finds that local meridional eddy MSE fluxes peak during cyclone intensification, particularly in the warm sector, and that a significant portion of these fluxes lies outside traditional cyclone masks. By including these external fluxes, the cyclone-attributed share of poleward MSE transport increases to about 30%. The analysis shows that cyclone characteristics (e.g., intensity, intensification rate) strongly influence the magnitude of attributed fluxes, both locally and in zonal integrals. On a seasonal scale, the most intense 25% of cyclones contribute nearly 45% of the cyclone-attributed AHT, despite being less frequent, while cyclone count alone is not a reliable predictor of seasonal flux. These findings highlight the dominant role of strong, rapidly intensifying cyclones in shaping transient eddy heat transport, while also acknowledging contributions from non-cyclonic or broader atmospheric features.
Despite the relevance and potential impact of the study, the manuscript is rather difficult to read. The presentation suffers from an excessive number of figures and appendices, many of which are hard to compare and detract from the clarity of the main message. The language is at times inappropriate or imprecise, requiring multiple readings of several sentences to grasp their meaning. This significantly undermines the accessibility and effectiveness of the work. Additionally, the text often repeats itself, rephrasing the same concepts unnecessarily, while methods are sometimes embedded within the results and discussions occasionally stray into irrelevant territory. These structural and stylistic issues pose a major obstacle to understanding and appreciating the main findings of this study. While I do see the merits of this work, which could be of interest to readers of Weather and Climate Dynamics, I cannot recommend it for publication in its current form. A major revision and restructuring are necessary. Below, I provide some general comments, followed by specific line-by-line comments to help the authors address the issues outlined above.
General comments
- You make use of many abbreviations, which in some points makes it quite difficult to understand what one is reading. I understand that some of these abbreviations are commonly used and the difficulty encountered might be subjective, still it would help if some of these were avoided. For instance, AHT might be easy to drop in favour of heat transport. You only talk about the atmosphere, so there’s no risk that it could get confused with ocean transport. Another example might be TE, which could be simply referred to as ‘transient’ as opposed to ‘stationary’ (SE).
- You also often directly use mathematical expressions (e.g., <v’m’_TE>, <v’m’ ^cycl _TE>, etc.) without giving them a descriptive name that would help the reader understand what they actually represent. For instance, <v’m’_TE> could be referred to as transient MSE flux, and <v’m’^cycl _TE> as the transient MSE flux attributed to cyclones, if I understood correctly.
- Concerning Figures, some panels could be better explained, as it is not always clear what different lines represent (e.g., see comment below on Figure 6).
- Figure B2 seems to me rather important, together with Figure 7, as they provide a visual example of your attribution method. As it is not straightforward to understand, please consider combining these figures and including the result in the method section. Furthermore, regarding Figure B2, it is not entirely clear to me yet what the difference between light and dark colour shading is. Does dark shading represent MSE fluxes that are attributed to cyclones?
- You measure the intensity of cyclones by looking at their minimum SLP, but at the same time acknowledge that in this way the analysis becomes biased towards high-latitude systems, given the climatological gradient in SLP (line 396). Why not considering a more objective measure of cyclone intensity, e.g. the maximum laplacian, which is not affected by the aforementioned bias?
- Regarding your attribution method, can you clarify how the MSE flux elements/patches are identified and partitioned? Is it using some kind of water-shedding algorithm?
Specific comments
Lines 1-23: The abstract could be a bit more concise and clearer; at a first read, there are a few statements that are not so easy to interpret (e.g., lines 18-20).
Line 22: the global warming statement is perhaps not so relevant here, I would consider removing it.
Line 40: why self-amplifying?
Line 42: largest anomaly based on what? Monthly, seasonally, yearly?
Line 46: Since “the” Earth
Line 52: In “the” midlatitudes
Line 54: not sure why you use a semicolon here
Lines 55—59: I am not sure I can fully follow these two statements, at least not without actually reading those references; can you try and be more explicit here?
Lines 55—71: In general, I feel like the introduction could start with this paragraph, where both background and aim for this study emerge more clearly. The previous two paragraphs are nice and true, but one may struggle to follow until the aim is revealed… in the third paragraph.
Line 72: I think you meant “concept”, not “conception”.
Lines 73-74: What do you mean by “circulation in the eddy”? I found the whole sentence not so easy to follow.
Line 81: what are these exceptions? Are there some data suggesting the number of cyclones is increasing?
Line 85: How is the Rossby radius of deformation expected to respond? Should it increase as opposed to what is observed, i.e. diabatically-driven cyclones becoming smaller under warming?
Lines 87—89: I am struggling to see how this statement is related to what comes before. You mean that part of the latent heat fluxes may be modulated by ocean dynamics and thermodynamics?
Lines 100-101: At a first read, the second research questions is not so clear. What do you mean by “cyclone characteristics […] reflected […] in the zonally integrated energy flux”? It might become clearer as one reads on, but so far I found it rather confusing.
Line 102: Why the focus on the seasonal scale? It comes a bit out of the blue.
Line 114: What are these levels?
Line 115: This statement is out of place here, so I would remove and keep it in the referenced section.
Line 126: 1 Bergeron corresponds to a drop of 24hPa in 24 hours, not 1 hour. This also means that if a cyclone minimum pressure drops by 1hPa in one hour, it does not necessarily correspond to 1 Bergeron, as the pressure drop must be calculated over a 24-hour period. Given the confusion, can you please clarify the intensification criterium by which you exclude cyclones?
Lines 128-129: I cannot understand this statement: how are leap days a problem? Also, don’t you focus on Souther Hemisphere winter, i.e. June, July, and August? There are no leap days there.
Lines 131-150: The atmospheric energy balance could be presented in a better way. Also, it is not clear from this paragraph why vertical averaged of m and v are subtracted before computing the meridional overturning circulation term in Eq. 2. Please clarify.
Line 157: Is 6PW referred to the maximum TE flux? Or is all at 45ºS? Why did you chose to highlight that latitude?
Line 158: You mean the method in Barpanda and Shaw (2017)? It is not clear why there would be discontinuities at the turn of each month, unless each month’s mean is removed rather than a centred 30-day mean. However, this is not specified anywhere. Please clarify.
Line 160: This sentence is a bit out of context in this paragraph. What methodological biases are averaged out by using a large cyclone sample size? Also, I feel like the use of cyclone tracks has not been properly introduced yet, making this statement quite obscure to interpret.
Line 161: ‘for to explore’, remove the ‘for’
Line 165: do you really need an abbreviation for high-pass?
Lines 191-203: Please consider avoiding the percentage and just use values between 0 and 1, as it makes formulas easier to follow (e.g., 2 x (1- p) )
Line 193: Any particular reason why this value was used?
Lines 205-211: This paragraph largely repeats what you say already in the methods section, consider removing it or highlight what new information is conveyed.
Line 211: What is “4 and 5”?
Lines 285-286: The sentence inside brackets does not really add much, please remove.
Line 288: Don’t you consider meridional fluxes by default when using v’? It is not clear to me what you mean to say with this sentence.
Line 312: Assuming ‘that’
Line 315: I don’t understand this sentence, in particular the ending ‘thereby the percentage to v’m’_TE’. Can you clarify?
Lines 316-322: I am not sure this paragraph adds much to the discourse. Either I am missing something or I would just remove it.
Line 323: What is ‘this latitudinal separation’? It is not so clear to me.
Lines 327-328: Maybe it is more a question of style, but I don’t think you need to anticipate what the next section is about at the end of each section. Similarly below, you repeat the finding of the previous section at the beginning of the new section. It reads somewhat redundant and does not improve clarity.
Line 332: The title is not so clear. What do you mean by ‘different’ cyclones? Is it based on intensity, duration, etc. In a way each cyclone is unique, so the title as it stands does not mean much.
Line 335: ‘the’ evolution
Line 338: I would remove the note ‘3’. A zonal mean is a zonal integral, so it does not need to be stated.
Line 346: Remove ‘than’
Figure 6: The way the caption is structured, it reads like the meaning of purple, black, and orange lines applies to panel (a) only. Please clearly state if the same holds for panels (b) and (c).
Line 354: No need to refer back to the methods section every time you use one of the methods, it is quite distracting.
Line 355: Here you only briefly refer to Figure 7, which is not described at all. It would actually be more appropriate to include it in the methods section, where you describe the attribution analysis.
Line 359: You mean an absolute increase? Because compared to the black line, the orange line appears to capture more than double the amount of MSE fluxes across all latitudes.
Line 364: Please use absolute latitude, as it is confusing to read ‘20º poleward’ when you actually are referring to a peak at 60ºS. Similarly below with ‘10º further equatorward’.
Line 367: Further equatorward than what latitude? In the subtropics there would not be any extra-tropical cyclones, which is what you focus on, so why would their absence in your tracks matter?
Line 376: Why are you citing those two papers here? Please remove as they are irrelevant here.
Line 397: Why 30h? I cannot understand this statement.
Line 403: Note 5 is redundant, you have already stated that you focus on different latitudinal bands, so they must be two different sets.
Line 409: The term ‘unjustly’ is not really appropriate, you mean systematically biased perhaps?
Line 413: Why is ‘total moisture’ italicised?
Lines 412-415: I struggle to understand the meaning of these two sentences.
Lines 416-421: This summarising paragraph is somewhat superfluous. Does it add anything to the previous ones? If not, either remove it, or integrate it in the conclusion section.
Line 425: Please check the grammar of this sentence, it reads a bit weird.
Lines 425-428: What are the other complementary perspectives presented in the previous sentence then? It is not really clear to me what you are trying to say in this paragraph.
Lines 430-431: Can be or are associated with? Also, what do you mean by ‘a latitude band of the order of 10º’?
Line 432: ‘This prompts the above framing that the averages…’ I really struggle to follow this sentence.
Lines 469-482: These paragraphs essentially explain your attribution method, thus it would be more appropriate to move them to the methods section, together with Figure 7.
Line 627: There seems to be some typos in the latex command for Stoll et al. (2023).
Citation: https://doi.org/10.5194/egusphere-2025-2314-RC3
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 422 | 95 | 15 | 532 | 37 | 49 |
- HTML: 422
- PDF: 95
- XML: 15
- Total: 532
- BibTeX: 37
- EndNote: 49
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1