Smoother versus sharper Gulf Stream and Kuroshio sea surface temperature fronts: effects on cyclones and climatology

Tsopouridis, Leonidas; Spengler, Thomas; Spensberger, Clemens

doi:https://doi.org/10.5194/wcd-2-953-2021

Articles | Volume 2, issue 4

https://doi.org/10.5194/wcd-2-953-2021

© Author(s) 2021. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/wcd-2-953-2021

© Author(s) 2021. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 2, issue 4

Research article

|

20 Oct 2021

Research article |

| 20 Oct 2021

Smoother versus sharper Gulf Stream and Kuroshio sea surface temperature fronts: effects on cyclones and climatology

Leonidas Tsopouridis, Thomas Spengler, and Clemens Spensberger

Download

Final revised paper (published on 20 Oct 2021)
Supplement to the final revised paper
Preprint (discussion started on 12 Oct 2020)

Interactive discussion

Status: closed

AC: Author comment | RC: Referee comment | SC: Short comment | EC: Editor comment

- Printer-friendly version

- Supplement

RC1: 'Review', Irina Rudeva, 10 Nov 2020
- AC1: 'Authors' response to Reviewer 1', Leonidas Tsopouridis, 19 Dec 2020
RC2: 'Review of Tsopouridis et al. (2020)', Anonymous Referee #2, 11 Nov 2020
- AC2: 'Authors' response to Reviewer 2', Leonidas Tsopouridis, 19 Dec 2020

Peer-review completion

AR: Author's response | RR: Referee report | ED: Editor decision

AR by Leonidas Tsopouridis on behalf of the Authors (20 Feb 2021) Author's response Author's tracked changes Manuscript

ED: Referee Nomination & Report Request started (24 Feb 2021) by Martin Singh

RR by Irina Rudeva (05 Mar 2021)

For final publication, the manuscript should be

accepted as is

accepted subject to technical corrections

accepted subject to minor revisions

reconsidered after major revisions

I would be willing to review the revised paper, if the editor considers it necessary.

I am not willing to review the revised paper.

rejected

Suggestions for revision or reasons for rejection

I appreciate the work the authors have done addressing previous comments. The writing has got clearer. I find it interesting that SST fronts do not seem to have a large effect on the characteristics of the cyclones over the Gulf Stream and Kuroshio. However, looking at Fig. 6c,d I can hardly see a difference in the SST between the control run and experiments despite the modified SST (Most likely it is the result of SST averaging in cyclone area though.)

My major concern is that the paper does not offer explanation as to why SST fronts seem to have little influence on cyclone characteristics. I would like to see if not a detailed analysis on this, but at least disucssion.

Despite the absence of an SST front in the Atlantic, the SST gradient over the black box remains strong compared to the rest of the ocean. Thus, even in the absence of a sharp SST front, this gradient may remain efficient in creating barcolinic instability. Jacobs et al. (2005) offered a very simple prestorm baroclinic index PSBI = (TG - Tland)/d, where d is the distance to the Gulf Stream from the coastline. The results shown in the paper make me think that land-sea contrasts might be significantly larger than the SST difference between CNTRL and SMTH experiments. I suggest calculating a similar index for cyclones in C1-C3 categories to check this (along which line this gradient should be calculated is an open question, e.g., along the same latitude as the cyclone centre at max intensity).

The other important question is why SST fronts play a larger role in the absence of cyclones. While land-sea contrasts, discussed above, can be particularly important in case of already formed cyclones, sharp SST fronts may play a large role during less perturbed flows, creating thermal wind and affecting the circulation downstream.

I am not sure if these suggestions are helpful, but the point is that I believe that the paper should offer a hypothesis that improve our understanding of the role of SST fronts in atmospheric dynamics.

Other comments (lines refer to tracked manuscript):

Title: climatology of what? I’d replace ‘climatology’ with ‘atmospheric flow’. The title still reads like a media headline rather than a name of a scientific paper, but I feel that this is the impression the authors wanted to make.

Am I right that cyclones selected for the analysis spent 3 6-hr steps and reached their max intensity within the black boxes? All other cyclones fell into category ‘absence of cyclones’?

l 537-8, 539: This is only true for cyclones that reach their max intensity over the Gulf Stream/Kuroshio, if I got it right. Cyclones downstream of the currents may still be modified by changing SST gradients (and baroclinicity) and play a role in forming those anomalies.

Why max wind speed is significantly displaced equatorward in the absence of cyclones but not when they are present? Is it because the jet is already displaced southward in both CNTL and SMTH simulations when a cyclone is present?

l. 245; Could slowing of the jet over the Gulf of Alaska along with increasing cyclone density be related to higher cyclolyse in that region?

Minor comments:

l.536: decrease in SST contrast
l. 537: Did you mean SST gradient, not just SST?

Fig. 4: I would suggest to show statistically significant differences. It is also interesting to see the trajectories of cyclones in supplements.

Fig.6: Is the SST averaged over 750km radius?

Fig.7-10: It is said that differences are statistically significant at 95% level. Is it true for all grid points even with differences close to zero? There are white areas in panels b and f, but not in c and g.

References:
Jacobs, N. A., Lackmann, G. M., & Raman, S. (2005). The Combined Effects of Gulf Stream–Induced Baroclinicity and Upper-Level Vorticity on U.S. East Coast Extratropical Cyclogenesis, Monthly Weather Review, 133(8), 2494-2501.

Hide

RR by Anonymous Referee #2 (08 Apr 2021)

For final publication, the manuscript should be

accepted as is

accepted subject to technical corrections

accepted subject to minor revisions

reconsidered after major revisions

I would be willing to review the revised paper, if the editor considers it necessary.

I am not willing to review the revised paper.

rejected

Suggestions for revision or reasons for rejection

Re-review of Tsopouridis et al. (2021).

I thank the authors for their efforts to address the reviewer concerns and in my opinion several aspects of the manuscript are much improved as a result. However, I feel that many of my comments were not properly addressed and I unfortunately remain skeptical of some of the analysis and conclusions. I recommend a major revision again, but I will leave future revisions up to the editor.

In order of response to reviewer:

- Please describe these papers correctly. The Parfitt and Seo (2018) paper indicates the importance of atmospheric fronts and the baroclinic waveguide (i.e. the combination of both cyclones and anti-cyclones), not extratropical cyclones specifically. This distinction is important. Also, many frontal detection algorithms detect atmospheric fronts in the Gulf Stream / Kuroshio regions without an “associated” extra tropical cyclone, especially with weaker fronts, as in Masunaga et al. (2020a, b). Simply including a larger cyclone radius does not address the issue of fronts discussed in previous work.

- The Bishop et al. (2017) paper is relevant as it shows a clear shift to ocean-driven variability. Numerous other papers show you need eddy-resolving SSTs to fully understand this variability and its influence on the atmosphere. e.g. Liu et al. (2021) - regardless of prescribed SSTs or not.
Liu, X., Ma, X., Chang, P., Jia, Y., Fu, D., Xu, G., ... & Patricola, C. M. (2021). Ocean fronts and eddies force atmospheric rivers and heavy precipitation in western North America. Nature communications, 12(1), 1-10.

-The above comment leads to the authors response regarding the lack of ocean eddy-resolving resolution in their simulations. I have two issues with their response. 1) It is not a valid argument to say that because the simulations and datasets have been used for studies previously, that they are suitable to use now. One would no longer solely use the CMIP2 models or ERA-40 to model climate variability, despite them being used extensively in the past. As mentioned above, many papers show lack of eddy-resolution in the ocean results in incorrect atmospheric responses. 2) In my opinion, papers like the one mentioned above clearly demonstrate the authors assertion that “it seems implausible that ocean eddies will fundamentally change… 1st order effect on synoptic-scale systems” is incorrect.

- I still do not believe that you can draw strong comparisons between the Gulf Stream and Kuroshio if the definitions used in each basin are different. I understand the magnitude of the SST gradients in the two basins are different, but I cannot accept the argument that this means it is impossible to come up with an SST front metric that fits both. This is also related to a comment I have on the SST gradient definition between CNTL and SMTHG/K (see below).

- The authors did not make an attempt to actually show how sensitive their results (ocean influence on atmosphere) in this paper are to cyclone detection. I would be surprised if the results were not sensitive to this, especially if the differences in "shallow and weak systems" are noticeable between algorithms as the authors mention.

- Related to my comment earlier, I still have a serious concern with using the SST front definition in the CNTL experiment to define the location in the SMTHG/K. The fact that no SST front is identified in the SMTHG/K with the current SST front definition simply tells me the definition needs to be altered. I suggest the authors review the numerous papers that have defined the Gulf Stream / Kuroshio location or the associated SST gradient in data as coarse as that used here. I have to say that this issue makes it extremely difficult for me to be confident in any of the CNTL vs. SMTHG/K comparisons. Especially as this makes the definitions for C1, C2, and C3 have little meaning.

- Regarding the logic behind only looking at cyclones undergoing maximum intensification right there, I understand the authors point that the maximum intensification of cyclones away from the SST front could not be directly associated with changes in SST. But that doesn’t at all mean that they can’t be. And in fact, it is fairly well understood that many ocean induced impacts are related to maximum intensification elsewhere (in which case you are not considering the whole set of scenarios). For example, changing the absolute SST will alter the magnitude and location of the evaporation input into the atmosphere. One can expect scenarios where the maximum impact of anomalous latent heating within the cyclone will occur downstream once moisture has risen within the system. This is also relevant for the authors response to a later comment (“ we thus believe that SSTs … are mainly important for climatologically setting the environment in which cyclones evolve, though each individual cyclone is not significantly affected directly “ ). What about an SST change that moistens the atmosphere prior to the cyclone arrival, but then the cyclone intensifies maximally downstream?

- For many of the reasons stated above, I still do not think the results in this manuscript can be used to claim that ’ SST fronts only have a minor impact on the characteristics and intensification of individual cyclones’. I would like to also point out that just because differences between CNTL and SMTHG/K are greater in some variables outside a cyclone radius, does not mean that SST fronts have a minor impact on individual cyclones.

Hide

ED: Publish subject to revisions (further review by editor and referees) (20 Apr 2021) by Martin Singh

AR by Leonidas Tsopouridis on behalf of the Authors (01 Jun 2021) Author's response Author's tracked changes Manuscript

ED: Referee Nomination & Report Request started (15 Jun 2021) by Martin Singh

RR by Irina Rudeva (24 Jun 2021)

For final publication, the manuscript should be

accepted as is

accepted subject to technical corrections

accepted subject to minor revisions

reconsidered after major revisions

I would be willing to review the revised paper, if the editor considers it necessary.

I am not willing to review the revised paper.

rejected

Suggestions for revision or reasons for rejection

I am pleased to see additional analysis on the role of land-sea contrasts vs SST gradients. In agreement with Reeder et al. (2021), the land-sea gradient, particularly in the North Atlantic, is the lead factor causing cyclo-/frontogenesis. Nevertheless, Reeder et al.’s results show that diabatic contribution modifiers the location of the maximum frontogenesis; however, this is less evident in the results of this paper (there is a larger difference in the Pacific though, fig. S3-4) . Perhaps, the model resolution is not enough to capture the affect of sharp SST gradients, while the large-scale SST gradients are not different between CNTRL and SMTH runs (i.e., between northern and southern boundaries of the selected regions), hence, as the authors had mentioned in their response, ‘the low-level baroclinicity (based on the temperature gradient at 850 hPa) remains largely unchanged’ compared to CNTRL run.

What remains puzzling is the equatorward shift of the jet; but I am happy for it to stay an open question. However, I still think the title of the manuscript is too provoking. The focus of the paper is on the cyclone response. The paper does not explain how localised changes in the SST gradients have such a significant impact on the atmospheric flow (or 'climatology') in the absence of cyclones - the changes are just a residual between total and when cyclones are present. If the residual is brought into the title, the study should address processes behind this in a bit more detail aiming to offer a plausible explanation of the mechanism leading to non-cyclone anomalies. Therefore, the title remains my main concern, but I am leaving it to the authors and/or the editor to decide if it is appropriate.

Minor comments (made to the version with tracked changes):

l.138: in line

l.221: perhaps, ‘elongated’ is better than ‘distributed’’

l.437: In my opinion, fig. S13 shows only marginal increase in the number of cyclones. Even if the difference is significant, it is not clear to me how it explains the equatorward shift of the jet.

L.514-515: I find this statement on the direct influence of the SST front to be very strong. As said in Reeder at al. ‘strong localised diabatic frontogenesis, which is amplified by adiabatic frontogenesis, can result in a front, which is consistent with atmospheric fronts in the region being most frequently located along the SST front’ - I fully agree with this statement and new plots (S12-13) make me think that a very weak effect on cyclones my be due to characteristics of the model. Also, in the phrase 'when no atmospheric fronts are present' replace 'fronts' with 'cyclones', as this paper does not explore atmospheric fronts

Fig. S12: It is hard to compare S11 and S12 by eye. Would be good if fig. S12 showed SMTH-CNTRL.

Hide

RR by Anonymous Referee #3 (07 Jul 2021)

ED: Publish subject to minor revisions (review by editor) (11 Jul 2021) by Martin Singh

AR by Leonidas Tsopouridis on behalf of the Authors (01 Sep 2021) Author's response Author's tracked changes Manuscript

ED: Publish subject to technical corrections (09 Sep 2021) by Martin Singh

AR by Leonidas Tsopouridis on behalf of the Authors (24 Sep 2021) Manuscript

Short summary

Comparing simulations with realistic and smoothed SSTs, we find that the intensification of individual cyclones in the Gulf Stream and Kuroshio regions is only marginally affected by reducing the SST gradient. In contrast, we observe a reduced cyclone activity and a shift in storm tracks. Considering differences of the variables occurring within/outside of a radius of any cyclone, we find cyclones to play only a secondary role in explaining the mean states differences among the SST experiments.