the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 23 Mar 2023
Internal tides off the Amazon shelf Part I : importance for the structuring of ocean temperature during two contrasted seasons
Abstract. Tides and internal tides (IT) in the ocean can significantly affect local to regional ocean temperature, including sea surface temperature (SST), via physical processes such as diffusion (vertical mixing) and advection (vertical and horizontal) of water masses. Offshore of the Amazon River, strong IT have been detected by satellite observations and well modelled ; however, their impact on temperature, SST and the identification of the associated processes have not been studied so far. In this work, we use high resolution (1/36°) numerical simulations with and without the tides from an ocean circulation model (NEMO). This model explicitly resolves the internal tides (IT) and is therefore suitable to assess how they can affect ocean temperature in the studied area. We distinguish the analysis for two contrasted seasons, from April to June (AMJ) and from August to October (ASO), since the seasonal stratification off the Amazon River modulates the IT’s response and their influence in temperature.
The generation and the propagation of the IT in the model are in good agreement with observations. The SST reproduced by the simulation including tides is in better agreement with satellite SST data compared to the simulation without tides. During ASO season, stronger meso-scale currents, deeper and weaker pycnocline are observed in contrast to the AMJ season. The observed coastal upwelling during ASO season is better reproduced by the model including tides, whereas the no-tide simulation is too warm by +0.3 °C for the SST. In the subsurface above the thermocline, the tide simulation is cooler by −1.2 °C, and warmer below the thermocline by +1.2 °C compared to the simulation without the tides. The IT induce vertical mixing on their generation site along the shelf break and on their propagation pathways towards the open ocean. This process mainly explains the cooler temperature at the ocean surface and is combined with vertical and horizontal advection to explain the cooling in the subsurface water above the thermocline and a warming in the deeper layers below the thermocline. The surface cooling induced in turn an increase of the net heat flux from the atmosphere to the ocean surface, which could induce significant changes in the local and even for the regional tropical Atlantic atmospheric circulation and precipitation.
We therefore demonstrate that IT, via vertical mixing and advection along their propagation pathways, and tides over the continental shelf, can play a role on the temperature structure off the Amazon River mouth, particularly in the coastal cooling enhanced by the IT. Keywords: internal tides, Amazon continental shelf and slope, temperature, modeling, satellite data, mixing, heat flux.
-
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
(3101 KB)
-
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(3101 KB) - Metadata XML
- BibTeX
- EndNote
- Final revised paper
Journal article(s) based on this preprint
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2023-418', Clément Vic, 06 Apr 2023
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-418/egusphere-2023-418-RC1-supplement.pdf
- AC1: 'Reply on RC1 : Clement Vic', Fernand Bernie Assene Mvongo, 21 Jul 2023
-
RC2: 'Comment on egusphere-2023-418', Nicolas Grisouard, 27 Apr 2023
Review of ‘Internal tides of the Amazon shelf. Part I’ by Assene et al.
This article investigates the role of tides, and in particular of internal tides, on the temperature field around the mouth of the Amazon river. The authors perform two NEMO multi-year runs, one with tides and the other one without. Whenever possible, they validate their results with observations, or a model fed by observations in the case of SSH. They then proceed to explain the differences in the temperature fields by carefully analysing various terms in the temperature evolution equation. In particular, after presenting the main features of the stratification and SSTs in each simulation over each analysis period (April-June and August-October), the authors investigate the role of a few proxies and processes to explain these features: air-sea heat fluxes; vertical temperature diffusion as a proxy for irreversible mixing; vertical temperature advection; and horizontal temperature advection.
The authors compare these metrics between two different seasonal averages: April-June and August-October, themselves averaged for three consecutive years. Their analyses also often focus on two transects that are roughly perpendicular to the shelf, along which elevated SSH variance indicates that the baroclinic tide has a greater amplitude there.
The upshot is that tides cause mixing along the thermocline, cooling the mixed layer and warming the sub-thermocline layer. Interactions with the atmosphere tends to damped the signal immediately under the surface, but said trend is strong around the thermocline. Horizontal and vertical advections of T play additional roles in setting the structure of the T field, although I have a few questions about this.
I have no reason to doubt the results of the paper, and the overall structure of the presentation is sound. I do have a few major comments about the presentation, albeit nothing that would put the publishability of the article in jeopardy. Within sections and paragraphs, the explanations are often hard to follow, sometimes because of long sentence structures, and sometimes because I could not distinguish between key points and details.
A. I was not able to wrap my head around the necessity to look at advection, at least in so much detail. I feel like the story is mainly in the mixing and heat fluxes. Advection seems to me that it is only here to balance the other irreversible processes. In other words, help me understand what I learn when looking at advection in such detail.
B. On this note, I don’t recall the authors explaining if the terms they investigate, the ones they deem most important, actually balance. Apologies if I missed it.
C. Often, figures are not cited in the right order. For example, panel (d) of figure X will be cited in the text before panel (a), when panel (a) is cited at all… If you could tighten up the organization of the figures (you could also consider removing a few panels here and there), it might make the flow of the article much smoother.
Here are medium-to-minor comments. Due to time constraints, I will not list English mistakes, typos and other minor sub-sentence edits I noted. Instead, I am attaching an annotated version of their pdf to this review. I want the authors to address my comments in this present review, but I do *not* need to see how the authors address my smaller comments on the pdf, even if they disagree with my suggestions. This would clutter the discussion too much.
1. Abstract is probably too long, be careful about word limits (and even if there aren’t any, trimming the abstract certainly wouldn’t hurt).
2. The ‘IT’ abbreviations: before streaming services, we used to listen to music on CDs. We withdraw money at ATMs. And we study the internal tide (IT), or we study internal tides (ITs).
3. All these parentheses in the middle of sentences make the text a little harder to read. You don’t need half of them.
4. Ll. 31-33: I am not sure what the sentence there means.
5. Ll. 61-62: ‘propagates it and dissipates it in the global ocean, *causing* diapycnal mixing…’. Diapycnal mixing is only a fraction of the energy dissipation.
6. Ll. 75-76: about the thermocline being a waveguide for internal tides. I do not remember Bordois’ work, but assuming my Ph.D. thesis wasn’t completely wrong, this statement is only true for high-frequency internal waves, typically internal solitary waves. Later, you observe significant mixing along the thermocline, so, this point might be relevant. But you don’t do anything with it, I don’t think you do a frequency analysis or attribute mixing to low-mode vs. ISW mixing. You could, which would require quite a bit of extra analysis. Or you could simply refrain from opening this can of worms…
7. L. 88: ‘and can thus modify temperature’: what is the connection with the beginning of the sentence?
8. L. 118: ‘(first 50 km)’: starting where?
9. L. 158: ISWs don’t really have wavelengths, which is a concept that applies to sinusoidal waves. You are probably thinking about the distance separating ISWs.
10. Ll. 249-250: ‘then tidal (…) all propagation’s modes’. I don’t understand this phrase.
11. L. 328: ‘The critical slope (…) 1.2’. I don’t understand this phrase.
12. In the same paragraph: careful about referring to Zaron’s product as ‘observations’. It is a product, derived from observations, via a model. He has many competitors, all of which produce quite different IT fields, even though they use the same data (see Carrère et al. 2021). So, it doesn’t have the ‘objective’ quality other observational products have, even though it’s all relative of course.
13. Ll. 340-341: ‘This longer period (…) observations.’ Except that Zaron’s product looks less smooth than yours… Maybe. Substantiate or drop.
14. Around the bottom of p. 11 and Fig. 2f: dissipation as I know it has one sign: all negative or all positive, depending on… well, whether you put a minus sign in front of the definition. Figure 2f shows a dissipation field that is mostly blue, but in some places, it is red. What does it mean for the baroclinic wave field to *gain* energy from dissipation?
15. L. 364, but also in other places: ‘simulation is about -1ºC cooler’: this might be nit-picky of me since we know what you mean, but the ‘-1ºC cooler’ could be understood as a double-negative, that is, ‘1ºC warmer’.
16. L. 379-380: did you ever explain why you chose AMJ and ASO over other periods? You compare AMJ with ASO (you do so in the abstract, by the way, which is not the appropriate place). But you didn’t explain why ASO was better than e.g. JFM. Note that in order to come to this conclusion, I did Ctrl+F on the abbreviations ‘AMJ’ and ‘ASO’, and there were no hits between the abstract and Section IV… If I’m wrong and you did explain it in one of Sections I to III (where it should be explained), then you need to introduce the abbreviations along with it. (This is one of the many instances where by suggesting something minor, I am actually hinting at a significant structural flaw in the paper… Often, information seems to be either missing or hard to find.)
17. L. 383: you reference Fig. 3e-g by itself, in section IV.1, which is different from where you referenced panels (a-d) (namely, III.2). This is a strong indication of either graphical information that should be broken up, or sections that should be consolidated. You could make those three panels a separate figure. While you could see this as a minor detail, Figure 3 is actually barely readable because everything is too small. And this is due to the 3x3 layout you chose, because you tried to show validation info (a-d, referenced in III) together with results info (e-g). Make (a-d) its own 2x2 figure, and make (e-g) another separate, 3x1 figure (recall that the final layout of Ocean Sciences is two-column).
18. Fig. 4c-d referenced on l. 389, before Fig. 4a (l. 399). If you feel the need to cite things in the wrong order, it is usually a sign that the overall results presentation needs to be re-thought.
19. L. 396: ‘Q_T’: isn’t it Q_t?
20. L. 410: Fig. 9f doesn’t exist.
21. L. 429: at this point, I started noticing that your reference temperature for what is a ‘cold’ isopycnal keeps fluctuating between 27ºC and 27.6ºC (which sounds dreamy, actually) throughout the article. You might want to decide on a choice earlier, in the methods section for example, and ideally use one value throughout. The ripple effect on the overall presentation might be significant.
22. L. 439: did you even define what the ‘mixing layer [depth?]’ was? NEMO product of dynamics-based?
23. Section IV.3, Vertical structure of the Temperature along A: You mention that ITs deepen the thermocline. But you only show results along A, which is a relatively narrow beam in space… Do I have to imagine the thermocline forming narrow ‘trenches’ along the beams, or is it a more global result? In other words, if you plotted a map of the termocline depth, would I see large deviations along the beams, or would I see a uniform deepening of the thermocline? If the former is true, plotting that map could be very nice (even better if it correlates with the dissipation pictures that you show later after). If the latter is true, you should mention it, because your choice of showing the results only along A goes against the message that the deepening is uniform.
24. Ll. 493-494: ‘the tidal simulation shows a decrease of the ZDF along the coast’. Decrease compared to what? Or going from which end of the coast to which end?
25. Ll. 495-496: This is not a true sentence, or not how to use ‘While’. Also, ‘almost closed’ [sic] is redundant. Same remark on l. 532.
26. Section V: I agree with Dr. Vic’s opinion that this section should be tightened to highlight key points. In general, I agree with his comments, by the way.
27. L. 728: ‘(…) of the PhD thesis of Fernand Assene (…)’
28. Fig. 1: Generation site letters A to F are hard to read, especially the black ones
29. Fig. 3: fonts are too small, among other things (see comment #17)
30. Fig. 6b: did you ever comment about the swirling, filamentary structures you see in the north-west corner of the figure? These are absolutely striking, but I don’t remember you commenting on them in the text. If you did, you might want to emphasize it a little better.
31. Figs. 6a and (especially) 6b: you mentioned at some point that you don’t focus on path B because it is similar to path A… but these figures, esp. (b), seem to indicate otherwise.- AC2: 'Reply on RC2 : Nicolas Grissouard', Fernand Bernie Assene Mvongo, 21 Jul 2023
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2023-418', Clément Vic, 06 Apr 2023
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-418/egusphere-2023-418-RC1-supplement.pdf
- AC1: 'Reply on RC1 : Clement Vic', Fernand Bernie Assene Mvongo, 21 Jul 2023
-
RC2: 'Comment on egusphere-2023-418', Nicolas Grisouard, 27 Apr 2023
Review of ‘Internal tides of the Amazon shelf. Part I’ by Assene et al.
This article investigates the role of tides, and in particular of internal tides, on the temperature field around the mouth of the Amazon river. The authors perform two NEMO multi-year runs, one with tides and the other one without. Whenever possible, they validate their results with observations, or a model fed by observations in the case of SSH. They then proceed to explain the differences in the temperature fields by carefully analysing various terms in the temperature evolution equation. In particular, after presenting the main features of the stratification and SSTs in each simulation over each analysis period (April-June and August-October), the authors investigate the role of a few proxies and processes to explain these features: air-sea heat fluxes; vertical temperature diffusion as a proxy for irreversible mixing; vertical temperature advection; and horizontal temperature advection.
The authors compare these metrics between two different seasonal averages: April-June and August-October, themselves averaged for three consecutive years. Their analyses also often focus on two transects that are roughly perpendicular to the shelf, along which elevated SSH variance indicates that the baroclinic tide has a greater amplitude there.
The upshot is that tides cause mixing along the thermocline, cooling the mixed layer and warming the sub-thermocline layer. Interactions with the atmosphere tends to damped the signal immediately under the surface, but said trend is strong around the thermocline. Horizontal and vertical advections of T play additional roles in setting the structure of the T field, although I have a few questions about this.
I have no reason to doubt the results of the paper, and the overall structure of the presentation is sound. I do have a few major comments about the presentation, albeit nothing that would put the publishability of the article in jeopardy. Within sections and paragraphs, the explanations are often hard to follow, sometimes because of long sentence structures, and sometimes because I could not distinguish between key points and details.
A. I was not able to wrap my head around the necessity to look at advection, at least in so much detail. I feel like the story is mainly in the mixing and heat fluxes. Advection seems to me that it is only here to balance the other irreversible processes. In other words, help me understand what I learn when looking at advection in such detail.
B. On this note, I don’t recall the authors explaining if the terms they investigate, the ones they deem most important, actually balance. Apologies if I missed it.
C. Often, figures are not cited in the right order. For example, panel (d) of figure X will be cited in the text before panel (a), when panel (a) is cited at all… If you could tighten up the organization of the figures (you could also consider removing a few panels here and there), it might make the flow of the article much smoother.
Here are medium-to-minor comments. Due to time constraints, I will not list English mistakes, typos and other minor sub-sentence edits I noted. Instead, I am attaching an annotated version of their pdf to this review. I want the authors to address my comments in this present review, but I do *not* need to see how the authors address my smaller comments on the pdf, even if they disagree with my suggestions. This would clutter the discussion too much.
1. Abstract is probably too long, be careful about word limits (and even if there aren’t any, trimming the abstract certainly wouldn’t hurt).
2. The ‘IT’ abbreviations: before streaming services, we used to listen to music on CDs. We withdraw money at ATMs. And we study the internal tide (IT), or we study internal tides (ITs).
3. All these parentheses in the middle of sentences make the text a little harder to read. You don’t need half of them.
4. Ll. 31-33: I am not sure what the sentence there means.
5. Ll. 61-62: ‘propagates it and dissipates it in the global ocean, *causing* diapycnal mixing…’. Diapycnal mixing is only a fraction of the energy dissipation.
6. Ll. 75-76: about the thermocline being a waveguide for internal tides. I do not remember Bordois’ work, but assuming my Ph.D. thesis wasn’t completely wrong, this statement is only true for high-frequency internal waves, typically internal solitary waves. Later, you observe significant mixing along the thermocline, so, this point might be relevant. But you don’t do anything with it, I don’t think you do a frequency analysis or attribute mixing to low-mode vs. ISW mixing. You could, which would require quite a bit of extra analysis. Or you could simply refrain from opening this can of worms…
7. L. 88: ‘and can thus modify temperature’: what is the connection with the beginning of the sentence?
8. L. 118: ‘(first 50 km)’: starting where?
9. L. 158: ISWs don’t really have wavelengths, which is a concept that applies to sinusoidal waves. You are probably thinking about the distance separating ISWs.
10. Ll. 249-250: ‘then tidal (…) all propagation’s modes’. I don’t understand this phrase.
11. L. 328: ‘The critical slope (…) 1.2’. I don’t understand this phrase.
12. In the same paragraph: careful about referring to Zaron’s product as ‘observations’. It is a product, derived from observations, via a model. He has many competitors, all of which produce quite different IT fields, even though they use the same data (see Carrère et al. 2021). So, it doesn’t have the ‘objective’ quality other observational products have, even though it’s all relative of course.
13. Ll. 340-341: ‘This longer period (…) observations.’ Except that Zaron’s product looks less smooth than yours… Maybe. Substantiate or drop.
14. Around the bottom of p. 11 and Fig. 2f: dissipation as I know it has one sign: all negative or all positive, depending on… well, whether you put a minus sign in front of the definition. Figure 2f shows a dissipation field that is mostly blue, but in some places, it is red. What does it mean for the baroclinic wave field to *gain* energy from dissipation?
15. L. 364, but also in other places: ‘simulation is about -1ºC cooler’: this might be nit-picky of me since we know what you mean, but the ‘-1ºC cooler’ could be understood as a double-negative, that is, ‘1ºC warmer’.
16. L. 379-380: did you ever explain why you chose AMJ and ASO over other periods? You compare AMJ with ASO (you do so in the abstract, by the way, which is not the appropriate place). But you didn’t explain why ASO was better than e.g. JFM. Note that in order to come to this conclusion, I did Ctrl+F on the abbreviations ‘AMJ’ and ‘ASO’, and there were no hits between the abstract and Section IV… If I’m wrong and you did explain it in one of Sections I to III (where it should be explained), then you need to introduce the abbreviations along with it. (This is one of the many instances where by suggesting something minor, I am actually hinting at a significant structural flaw in the paper… Often, information seems to be either missing or hard to find.)
17. L. 383: you reference Fig. 3e-g by itself, in section IV.1, which is different from where you referenced panels (a-d) (namely, III.2). This is a strong indication of either graphical information that should be broken up, or sections that should be consolidated. You could make those three panels a separate figure. While you could see this as a minor detail, Figure 3 is actually barely readable because everything is too small. And this is due to the 3x3 layout you chose, because you tried to show validation info (a-d, referenced in III) together with results info (e-g). Make (a-d) its own 2x2 figure, and make (e-g) another separate, 3x1 figure (recall that the final layout of Ocean Sciences is two-column).
18. Fig. 4c-d referenced on l. 389, before Fig. 4a (l. 399). If you feel the need to cite things in the wrong order, it is usually a sign that the overall results presentation needs to be re-thought.
19. L. 396: ‘Q_T’: isn’t it Q_t?
20. L. 410: Fig. 9f doesn’t exist.
21. L. 429: at this point, I started noticing that your reference temperature for what is a ‘cold’ isopycnal keeps fluctuating between 27ºC and 27.6ºC (which sounds dreamy, actually) throughout the article. You might want to decide on a choice earlier, in the methods section for example, and ideally use one value throughout. The ripple effect on the overall presentation might be significant.
22. L. 439: did you even define what the ‘mixing layer [depth?]’ was? NEMO product of dynamics-based?
23. Section IV.3, Vertical structure of the Temperature along A: You mention that ITs deepen the thermocline. But you only show results along A, which is a relatively narrow beam in space… Do I have to imagine the thermocline forming narrow ‘trenches’ along the beams, or is it a more global result? In other words, if you plotted a map of the termocline depth, would I see large deviations along the beams, or would I see a uniform deepening of the thermocline? If the former is true, plotting that map could be very nice (even better if it correlates with the dissipation pictures that you show later after). If the latter is true, you should mention it, because your choice of showing the results only along A goes against the message that the deepening is uniform.
24. Ll. 493-494: ‘the tidal simulation shows a decrease of the ZDF along the coast’. Decrease compared to what? Or going from which end of the coast to which end?
25. Ll. 495-496: This is not a true sentence, or not how to use ‘While’. Also, ‘almost closed’ [sic] is redundant. Same remark on l. 532.
26. Section V: I agree with Dr. Vic’s opinion that this section should be tightened to highlight key points. In general, I agree with his comments, by the way.
27. L. 728: ‘(…) of the PhD thesis of Fernand Assene (…)’
28. Fig. 1: Generation site letters A to F are hard to read, especially the black ones
29. Fig. 3: fonts are too small, among other things (see comment #17)
30. Fig. 6b: did you ever comment about the swirling, filamentary structures you see in the north-west corner of the figure? These are absolutely striking, but I don’t remember you commenting on them in the text. If you did, you might want to emphasize it a little better.
31. Figs. 6a and (especially) 6b: you mentioned at some point that you don’t focus on path B because it is similar to path A… but these figures, esp. (b), seem to indicate otherwise.- AC2: 'Reply on RC2 : Nicolas Grissouard', Fernand Bernie Assene Mvongo, 21 Jul 2023
Peer review completion
Journal article(s) based on this preprint
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 405 | 243 | 24 | 672 | 17 | 15 |
- HTML: 405
- PDF: 243
- XML: 24
- Total: 672
- BibTeX: 17
- EndNote: 15
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Cited
1 citations as recorded by crossref.
Ariane Koch-Larrouy
Isabelle Dadou
Michel Tchilibou
Guillaume Morvan
Jérôme Chanut
Vincent Vantrepotte
Damien Allain
Trung-Kien Tran
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(3101 KB) - Metadata XML