Review of “Synoptic observation of full mesoscale eddy lifetime and its secondary instabilities in the Gulf of Mexico” by Charly de Marez.

This is a concise, well written paper summarising observations of a loop current eddy’s life cycle in the Gulf of Mexico using high resolution altimetry from the relatively recently launched SWOT satellite.  Direct observation of key dynamical processes such as eddy shielding, high-mode instabilities, and dipolar interactions were made possible by the resolution and coverage of the SWOT data.  Although short and highly focused, this paper provides significant insight into processes which likely affect eddies throughout the global ocean.  Some minor improvements to the text and figures are needed, but once completed I would recommend this paper for publication.

Comments on figures

All figures: the Ocean Science style guide says panel labels should be enclosed in brackets on the figure.

Figure 1:

I can see why you’ve chosen to display this the way you have – you get all of it on one figure, and the days you want to emphasise are larger.  However, I think skipping between the larger and smaller panels does disrupt the flow for the reader because now you don’t just start at top left, work across the row, go down one row, work from left to right, etc.  Given that you currently only have two figures in the whole paper, I suggest you consider stretching this out - perhaps three figures each with six panels of equal size, for example?

Are you sure the colour scale is colourblind-accessible?  And personally, I find diverging colour scales which aren’t centered at zero rather confusing.

I think the AVISO SLA is maybe a bit too pale.  For example, I really can’t see LCT₁ on panel m, it just looks white.  You could make the AVISO SLA quite a bit less pale while still maintaining a good contrast between it and the SWOT SLA.

Panels b, c, d, e, h, I, j, k, n, o, p, q have text above them which is far too small to read – I had to zoom in to 200% to see it was their dates.  Since they’re not all regularly spaced in time, the reader needs to be able to see the dates.  These panels also lack axes labels, and panels n and p have arrows which are not explained in the caption nor explicitly mentioned in the main text.

The labels LCE₀, LCT₀, LCE, VS, C₁, C₂, C₃, C₄ and LCT₁ are not explained in the caption, nor have they been mentioned in the main text at the point where you first refer to this figure.  I see that there’s a lot of explanation in the main text of these labels, and it’s understandable that you don’t want to put all of this in the caption, but perhaps you could say something like “The labels LCE₀, LCT₀, LCE, VS, C₁, C₂, C₃, C₄ and LCT₁ will be discussed in the main text.”

You don’t need the colourbar three times on one figure, but you do need to label it.

Figure 2:

On figure 1, you said “The black contour marks the LCE detection from gridded altimetry, while thin gray contours represent iso-SLA lines from SWOT passes at 5 cm intervals”.  Here on figure 2 you say “Black contours indicate iso-SLA lines from SWOT passes at 2 cm intervals”, but it looks like you still also have the black contour which is the LCE detection from gridded altimetry, even though this isn’t mentioned in the caption.

Panels a and d – is this exactly the same colour scale as in figure 1?  It worth be worth either saying so in the caption, or altering the colourbar so it has the same tick marks as in figure 1, which will make it more obvious that it’s the same scale.  In any case, the AVISO SLA is clearly not as pale here as on figure 1, and it would be better if the two figures were consistent.

Re the black contours which indicate iso-SLA lines from SWOT passes at 2 cm intervals – it looks like you’re only showing these in certain areas, i.e., outside the contour marking the LCE detection from gridded altimetry, but this isn’t stated in the caption.

Panels b and e show “Geostrophic current magnitude derived from the denoised SWOT SLA”, but you clearly have data outside the SWOT swaths.  Ditto panels c and f.

In figure 1 your panel labels went left to right and then down to the next row.  In figure 2 your panel labels go down the first column and then down the second column.  It’s easier for the reader if you do them the same way in all figures.  Ditto figure B1.

Panels b to f have no axes labels or visible tick marks.  I think there are grid lines on all panels but they’re so faint they’re very difficult to see.  Please make them like figure 1’s grid lines, and add axes labels.  You don't need axes labels on all panels but you do need x-axis labels on the bottom row and y-axis labels on the left-hand column.

Figure B1:

The colourbars and length scale are very difficult to see, even zoomed in.  Please put them outside the panels.

The text at the top of each panel is also quite small and hard to read.

The dashed and solid contours on panels e and f are difficult to see – maybe use a contrasting colour?

Textual comments

Line 52: “This newly formed eddy was then trapped within a train of alternating-polarity eddies (Figs. 1d-f).”  To my eyes, Fig 1d doesn’t look that different to panels b or c, so it’s not clear to me at what point you would start describing something as “trapped within a train of alternating-polarity eddies”.

Line 128: Seagliders is a specific brand name, the generic term is ocean gliders.

Line 168: sutructures

Line 171: which gridded altimetric product?  Even if you don’t want to show it, you could name it.

Line 173:  SWOT measurements are not instantaneous.  It takes time for the satellite to orbit around the globe.  Perhaps you meant that the measurements are near-instantaneous compared to timescales of interest?

Line 174: “time gaps between passes remain too small to affect our conclusions.”  Could you say a little more here about the timescales of interest?  It is clear from figure 1 that changes do occur on weekly timescales.

Line 177:  Fig 1l looks a lot like figure eleven in this journal’s typeface – you could miss the letter l out of your panel labels and just go j, k, m.

Line 188: “simulations are ran over a year.”  This is bad grammar, please correct.

Line 190:  “Timesteps are adjusted to respect the CFL criterion.”  You don’t say what this criterion is, or what CFL stands for, or provide a reference.

Line 191: “the smallest as possible” – either “the smallest possible” or “as small as possible”.  Plus “similarly as in” is bad grammar.  Perhaps just “as small as possible (Callies et al., 2016).”

Line 192:  delete “similarly”.

Line 204:  “We chose R = 100km” because? 

This paper describes the sea surface height and vorticity structure at the high resolution obtained by SWOT for the case of one Loop Current Eddy, from the moment of its formation to part of its evolution as it drifts westwards. It is certainly nice to see how the promise of SWOT actually looks like in this region after all these years of research using numerical models and sparce data to get an idea of what these measurements may be able to resolve.

Nevertheless, I consider that given the information provided many statements and conclusions need to be toned down, particularly in Section 2 and the Conclusions, or in some cases elaborating a little more on how the conclusions are reached is necessary. I  provide specific examples below. 

Another general comment is that it would be nice to have include more references that back up or provide more guidance to the reader in the Introduction (below I point to specific cases), particularly with respect to the “theoretical and numerical understanding of mesoscale eddy dynamic… offering insights into eddy behavior.” so as to have a better idea of the theoretical and conceptual framework, and the details they entail. Elaborating a little on that would also be helpful.

The numerical results that back up what is observed are certainly a very nice complement, but are barely described.  I would find a much more convincing case that what is observed with SWOT is that particular eddy instability with a bit more detailed comparison. For example,  do the 4 cyclones so formed around the LCE evolve in aproximately the same time scales (the observations show that 4-mode state lasts just a couple of weeks, how do these timescales  compare in terms  of, say, number of inertial periods in the simulations and in the LCE)? How many inertial periods does the eddy last from the moment the instabilities appear and its destruction, and how does that time compare to the period the LCE was followed? Do you see such things such as migration away from the eddy (C1 and C2)  or merging to evolve into the dipolar structure that is associated to the westward migration of the eddy? 

Section 3 I find no large objections to since indeed this is where the SWOT resolution certainly shows something that was not easily observed before around LCEs. It would be nice though if something is said on how much we can trust the vorticity obtained from SLA at those small scales and high Rossby numbers (i.e. geostrophic currents from SLA slopes), see for example Tranchant et al 2025.  Although I do believe this is the first time I see in an article the observations of SWOT on a LCE, there certainly is more literature  where SWOT has been used to reveal  submesoscale features surrounding mesoscale fields with some in depth analysis on how much certain processes are represented (e.g. Agulhas retroflection in Coadou-Chaventon 2025). These are all very recent and may not have been available to the author when he submitted, but I think some of those results may help make a stronger case for what is being stated here with only a few snapshots of the SWOT derived fields. Including such references at least in the ending statement of the Conclusions would certainly help to make a more informative and stronger case for SWOT’s potential to solve all that is mentioned there. 

Detailed comments

Abstract

Maybe tone down a  bit “this has not been previously characterized but in theoretical or numerical models”. At least dipolar interactions (see figures later on) can be observed even with the low res altimetry products and other observations prior to SWOT. 

L 20. some citations would be helpful.

L 45. Previous sentence talks of all LCEs, this sentence implies it talks of a specific one. Maybe repharse to say this will be possible to be done to LCEs present in that period of time.

L 51. Can you provide reference with an example of a reatachament observed with gridded altimetry reattachment?

L 54. Note that the vorticity shield is not really shown but rather inferred from the shape of the eddy. I am not convinced the cyclonic vorticity ring around the LCE is not observed with low res altimetry, how do you determine that eddy shielding can only be inferred from SWOT data, as claimed in the abstract?

L 55. Please provide reference defending that the difference in sizes of LCEs depends on if they are formed with and extended vs no extended Loop Current. In any case, maybe this info doesn’t matter much since nothing has been stated on why the size of the eddy would matter for the rest that is being discussed.

L57-80. I myself find it hard to tell if C1-4 where formed by eddy destabilization, given that cyclonic structures were already present surrounding the LCE0 formation, and may be precursors of those 4 cyclones associated to the “squared vortex”, and they last a very short period of time in that layout. Would be nice to see how the evolution fo the theoretical case shown in Appendix B looks in comparison as those cyclonic eddies evolve, see my general comment on that in the beginning of this review.

Another thing to point out is that, eventhough not as nicely, the low res altimetry does show that layout (see attached file). This may be not so surprising given that the cyclones that evolve from the instabilities are comparable in size to the original LCE according to the simulations.  I say this cause there are examples of LCEs that have evolved in a relatively quiet mesoscale background (Poseidon, Meunier et al. 2018; Kraken, Beron-Vera et al. 2018, see attached file)  which lasted a long time (months) without showing such strong deformations that could as clearly be attributed to eddy instabilities such as those in figure B1. Given that Poseidon was particularly large and energetic and does not seem to interact with that many other mesoscale structures present in the gulf, wouldn’t it have evolved instabilities with such  cyclones and be destroyed quicker? How comparable are the time scales in the simulation from the appearance of the cyclones and the destruction of the eddy to the lifetimes of Kraken and Poseidon?

Don’t get me wrong, I would more than happy be convinced, but as of now I don’t find evidence that support the following conclusions:

1. The eddy instability process can only be inferred by SWOT, or in other words, the low res altimetry data would not be able to provide hints of it. Particularly given that the cyclones that appear are of approximately the same scales as the LCE.  At least in the eddy in question, the low res does show hints of those cyclones

2. The description of the SSH evolution of figure 1 is proof that the LCE destabilized as expected from theory. At the most I would say this particular case invites to think of this instability process.

3. The eddy did not get destroyed by its own instability because “external conditions and interactions with surrounding eddies prevent the destruction of the mode-4 LCE, leading to a complex interplay of processes.” Can you describe more clearly how you unequivocably arrive at this conclusion?  

4. “The westward drift is not driven by planetary Rossby wave propagation, as traditionally assumed… instead, it is a direct consequence of the dipolar interaction”. How is that so clearly determined from these images? could it be both? Note that the dipole is clearly seen in the low res altimetry (images provided in attached file), while no such dipole is present for Poseidon nor Kraken, and they also migrated westwards.  From how all of this is phrased, it seems to imply that  LCEs migrate westwards because they are in fact dipoles, and those dipoles only exist because of the evolution of the cyclones that appear due to the eddy instability. But this is certainly not always the case.

L 94. Given all the above, “The present observation confirms…. “ and “fully consistent with theoretical and numerical "vortex studies”” seem to strong statements.

L 97. Maybe add what a vorticity shield refers to, if I understand it right it is a ring of opposite vorticity (cyclonic) surrounding the eddy.

L 100  “ unambigous evidence….” at least from the evidence provided in this study, seem too strong a statement here.

L 145  I would tone this down too,  since the eddy was still “alive” at the end of the study period. I did continue having quite an interesting evolution that can be traced by low res altimetry until at least July 2024 (see attached file).

Figure 1. Would be nice if labels for C1-C4 in panels m - r are included (merging also, maybe as C1+C2 or something like that) to better follow the sequence described in the text.

REFERENCES

Beron-Vera, F.J., Olascoaga, M.J., Wang, Y. et al. Enduring Lagrangian coherence of a Loop Current ring assessed using independent observations. Sci Rep 8, 11275 (2018). https://doi.org/10.1038/s41598-018-29582-5

Coadou-Chaventon, S., Swart, S., Novelli, G., & Speich, S. (2025). Resolving sharper fronts of the Agulhas Current Retroflection using SWOT altimetry. Geophysical Research Letters, 52, e2025GL115203. https://doi.org/10.1029/2025GL115203

Meunier, T., Pallas-Sanz, E., Tenreiro, M., Portela, E., Ochoa, J., Ruiz-Angulo, A., and Cusí, S.: The vertical structure of a Loop Current Eddy, J. Geophys. Res. C: Oceans, 123, 6070–6090, https://doi.org/10.1029/2018JC013801, 2018.

Tranchant, Y.-T., Legresy, B., Foppert, A., Pena-Molino, B., & Phillips, H. E. (2025). Swot reveals fine-scale balanced motions and dispersion properties in the Antarctic circumpolar current. Authorea Preprints.