the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 25 Jun 2025
Internal tide signatures on surface chlorophyll concentration in the Brazilian Equatorial Margin
Abstract. This study investigates the influence of tides on chlorophyll-a (CHL) variability in the North Brazilian Equatorial Margin using daily remotely sensed CHL data from 2005 to 2021. The impact of the tides is deduced by comparing the spring tides with the neap tide signal (fortnightly signal, 14.7 days) in the GlobColour and MODIS-AQUA spring-neap tide composites. Results show that, on the shallow Amazon shelf, significant fortnightly CHL variability is primarily driven by barotropic tide-induced friction on the shelf that produces a significant vertical mixing. On the northwestern shelf, where the Amazon River plume dominates, sediment resuspension (likely driven by stronger tidal mixing during spring tides) suppresses primary production. In contrast, the increased stratification of the river plume during neap tides enhances nutrient availability likely enhancing primary production, leading to negative spring-neap CHL differences (GlobColour: -50 %, MODIS-AQUA: -84 %). Conversely, the northeastern shelf, where low turbid waters are present, exhibits positive CHL differences (GlobColour: +30 %, MODIS-AQUA: +70 %), likely caused by nutrient-rich uplift. Offshore, baroclinic tides, also known as internal tides (ITs), induce CHL-positive spring-neap tide differences in the spring-neap tide composite with a spatial structure of a wave-like pattern along IT pathways. These anomalies are spaced by mode-2 wavelengths (about 68 km), with peak values reaching +3.3 % (GlobColour) and +9.0 % (MODIS-AQUA). The observed wave-like pattern may be attributed to two potential mechanisms. First, tide aliasing caused by the satellites’ orbit characteristics, which consistently capture IT wave crests at similar locations, with concurrent modulation of the deep chlorophyll maximum (DCM) due to IT wave passage along the thermocline. Second, the propagation of internal tides (ITs) as beams into the ocean interior (“Ray theory”), causing upward displacement of isotherms and driving nutrient fluxes that enhance primary production, supported by ray-tracing analysis. Wave patterns in CHL spring-neap tide composites from GlobColour and MODIS-AQUA suggest contributions from mode-1 and mode-2 internal tides. Results indicate also a lag of 1–3 days between spring-neap tides and peak chlorophyll variability, indicative of maximum mixing. The effects of ITs on CHL are more pronounced than on sea surface temperature, likely due to differences in sensor penetration depths and the influence of air-sea interactions.
- Preprint
(6216 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
- RC1: 'Comment on egusphere-2025-2307', Anonymous Referee #1, 01 Aug 2025
-
CC1: 'Comment on egusphere-2025-2307', Longyu Huang, 06 Aug 2025
In this manuscript, the authors investigate the influence of tides on chlorophyll-a (CHL) variability in the Brazilian Equatorial Margin using two types of daily remotely sensed CHL data from 2005 to 2021. Although the studies of internal tides (ITs) in this region have been widely documented, the ITs contributions on the marine ecology are rarely reported. This is an interesting topic and suitable for the journal Ocean Science. Overall, the current manuscript is well organized and written, the results are present clearly and provide valuable insights into the ecological impacts of ITs. Now, I recommend a minor revision and some questions and suggestions for the authors to improve the manuscript.
Specific comments:
- The introduction is well written and clear, and summarizes the relevant research on internal tides in this region. However, in the main text, the authors mention lots of elements about the bathymetry, generation sites and pathways of internal tides in the Brazilian Equatorial Margin that referred from other studies. I suggest the authors give a figure to present the overview for the study region.
- In Methods 2.2, why the average period is 15 days?
- In Methods 2.3, the physical mechanisms of wavelet analysis should be explained and say somewhat to explain the meanings of high or low Power in Figure 5 (a-b).
- In Methods 2.4, the tides in single ITs generation sites are extracted for the calculation of Tide Composition f (S, N). In Fig. 6-9, there are several ITs generation sites and pathways, such as pathway A, B and C, I wonder if the f (S, N) is computed using tides extracted in one site or all sites?
- In Figure 2 and 3, I suggest the author added the bathymetric contours for the clarity. Besides, the locations of points A and D are hard to distinguish, the point color should be changed. What the two black dashed lines in the lower left mean? The locations of mode-1 and 2 ISWs should be labeled in Figure 2 and 3 (d).
- Line 210: Which IT generation site should be clarified.
- Line 213: Figure 4 show that signal filtered mode-2 IT is more closely with the original signal of CHL different. If this mean that the variation of CHL is mainly caused by mode-2 IT? I wonder why the smaller horizontal and vertical scale of mode-2 IT could induce greater variation of CHL.
- Figure 5: the barotropic and baroclinic energy flux (in c and d) are from NEMO by Assene et al 2024, if the time align with (a) and (b)? Besides, in Line 225, why choose S2 tidal constituent instead of M2?
- Line 239: where are the three peaks of positive CHL difference?
- Line 242: Figure 7?
- The authors should explain why the mode-2 f is much greater than mode-1 f in Figure 9, while the values of f are equivalent in Figure 7, even though two types of CHL data are used.
- The display range of the color bar should be uniform. Such as (a-f), (g-h) in Figure 6 and 8, respectively, and Figure A1 (a-b).
Citation: https://doi.org/10.5194/egusphere-2025-2307-CC1 -
RC2: 'Comment on egusphere-2025-2307', Anonymous Referee #2, 02 Sep 2025
The manuscript (MS) makes a comprehensive and detailed analysis to connect wave-like signal in chlrophyll-a satelitte images to internal tide dynamics off the Amazon continental shelf. Although the title and abstract suggest a focu on baroclinic tides, the MS also presents ideas connecting observed patterns to barotropic tides. The results are interesting and certainly relevant, but the MS requires extensive reorganization for the main message to come through clearly. The knowledge gap is not well-defined, and in several parts of the document, the arguments are largely speculative and not sufficiently supported by evidence. I recommend major revision. If the speculative aspects are clarified, supported by evidence, or removed, and the MS is restructured, it could make a strong contribution to the literature.
MAJOR CONCERNS
My major concerns relate primarily to structure and terminology:
1. Structure of Introduction and Conclusion:
The introduction reads almost like a discussion. Although the last paragraph mentions the topic of the MS, the knowledge gap or main question is uncluear.
Similarly, the conclusion reads like a dicussion and is repetitive compared to Section 4.3. I sugget renaming it Summary and Conclusions.
2. Terminology:
The use of showcase seems inappropriate in this context. Consider using case study or illustrative case.
Sunglint should not be used as an adjective.
Although I think I understand the reason behind the two study cases, the MS should clarify why they are used. It should explicit that they are intended to demonstrate the existence of internal tides in the area, which justifies exptrapolating to a time series of chlorophyll images from Globcolour and MODIS-Aqua.
Avoid using coloquial terms like “sandwich”; I recommend “to be flanked” instead.
The term “spring-neap tide” is confusing; I believe you mean “spring-neap tidal cycle”. Clarify or define this early in the MS.
Please correct the spelling of “MODIS-Aqua” consistently throughout the MS.
SPECIFIC COMMENTS (line numbers are referenced)
51: the availability of PAR.
53: Define interfacial ITs.
55: Use deep or subsurface chlorophyll maximum
61-62: “to two combined effects”
72: Indonesian Seas (there are several of them) or Indonesian Throughflow
117: Suggestion: more than 20% of the time series.
129: Define what is a showcase day.
163: Is this assumption valid for this region? The freshwater inflow could have drastic effects on local stratification, likely preventing gradual and continuous profiles. How might this assumltion affect your analysis? You should demonstrate the valifity of your asumption specially becasue you are looking into a surface signature of these waves.
173: N (or M – the lateral buoyancy gradient) will certainly exhibit lateral variability due to riverine input and submesoscale/mesoscale stirring.
Figure 2: Points A and B are barely visible. The magenta font over red/orange background lacks contrast. Also, clarify which quantity is shown in panel D? In addition, having the coastline would improve visualization, orientation and highlight the relevant features. ISW signatures do not stand out, significant zooming is needed to see them.
189, 192: References for the typical waelengths are needed.
205: Clarify whether the average is temporal only (one value) or also considers spatial differences (2D map).
Figure 4: Consider adding arrows to highlight peaks.
216: statistically significant.
249: Difference relative to what? Please clarify.
337-341: Here and later on. How would the DCM be visible from remote sensing? Does IT-induced mixing bring chlorophyll from the DCM to the surface?
379: The MS focuses on internal tide signatures in surface chlorophyll; clarify why barotropic tides are also discussed.
386-389: These statements are speculative. Explain more clearly how the time series analysis supports these conclusions.
389: Clarify “lower turbid”?
392-393: “as they propagate offshore the open ocean”, likely a typo; please check.
402: […] beam signal is attenuated after 300-400 km […]
404: more effective TO OBSERVE.
417: How is stratification a “background circulation feature”?
421-422: This statement is clearer than Lines 386-389.
423-424: Could the difference between tracers simply be because chlorophyll is biologically reactive while temperature is a physical tracer? The current explanation is overly complicated. I do not agree with your argument.
Citation: https://doi.org/10.5194/egusphere-2025-2307-RC2
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 492 | 64 | 14 | 570 | 13 | 28 |
- HTML: 492
- PDF: 64
- XML: 14
- Total: 570
- BibTeX: 13
- EndNote: 28
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Merit:
The manuscript is concise and well written. I think the material would fit well in EGU Ocean Science and I anticipate it ultimately will be a nice contribution. While the figures and results are mostly convincing, I think there are a few weak points in the presentation and analysis, which I have mentioned below.
Major comments (see line-by-line comments for more details):
Line-by-line comments and suggestions:
L8 – Perhaps it would be better to state that “chl was higher during neap tides”, etc. rather than the “chlorophyll differences” term which could be confusing (in the abstract only).
L11 – clarify what the composite means if you mention in the abstract
L19 – Clarify that this means chl max/min values occur 1-3 days after the tidal signal
L32 – I would recommend to add a bit more detail on how season conditions vary and how this modulates ISWs
L56 – “remote sensing”?
L71 – Do you mean “intra-annual” (or if not, please explain further)? As written it implies variability over many years, and I don’t understand how that would be impacted by tides.
L94 – Please explain why those two days were chosen. Were they objectively chosen? Are they representative of commonly observed conditions? I think it’s still probably ok if not, but in that case it should be more clear if these may not be typical of what’s usually observed.
L117 – I am concerned that such an exclusion will be seasonally dependent; i.e., discharge of the Amazon, and presumably turbidity as well, will vary seasonally. Thus, the excluded data will be disproportionally from a certain time of year. Please explain/quantify whether this is the case, and how this choice may have influenced later analysis and results.
L149 – I don’t understand exactly what S and N are. Are they the concentration of CHL corresponding to the center of the spring/neap time or something else?
L155 – What do you mean by “composite maps”? I don’t see any spatial term in Equation 3. From the equation it looks to me like the lagged dependence of chl on tidal cycles. Is this applied to different points in space? I am probably missing something here; please explain this in more detail.
Reading on, I can see the figures of this metric. I still think more explanation here would be helpful.
L163 – I do not think these stratification assumptions are valid in the region of study. Because of the Amazon plume, I suspect that the profile of N is highly variable and not gradual nor continuously varying (you seem to also mention this at L173). Are there any references that show how much N varies? I think some measure of uncertainty should be included and quantified.
Fig 2 – I would recommend to include in the caption or figure what “difference” specifically refers to.
Fig 2 a/b (and other figures) – I would suggest to change the color scale. Rainbow color scales are not perceptually uniform, and thus the magnitude of features such as fronts may be enhanced/biased by the scale. I would recommend to use “cmocean” or a similar perceptually uniform scale.
L213 – For clarity, the point here is that the mode 2 filtered signal is strongly correlated with the chlorophyll difference, correct? It might be helpful to quantify this correlation in some way.
L215/Fig 5 – There needs to be a physical explanation of what the result of the wavelet analysis is showing. It is just the spectral energy at a period of 15 days so an indicator of regions where spring/neap tides are strongest, correct? Or maybe I am missing a bit of it?
L252 – It would be helpful to quantify the coherence of the signals for each of the pathways. I agree with the findings visually, but having a quantitative comparison would make your argument stronger.
Fig 7 – This figure only shows lags up to 3 days. I would include days 4 and 5; otherwise it is unclear that there is not an even stronger signal with more time lag.
L321 – I’m still not quite convinced that aliasing does not significantly impact these results; I suspect there will be regional differences from the North Sea to the area of study. Are there other references that have looked at this in similar conditions (and found that the bias was between flood/ebb and not high/low)? In any case, I think more explanation beyond citing a single study would be helpful.
L350 – These factors would be expected vary seasonally. I would recommend to discuss in a bit more detail what conditions were present in the two showcase times. Regarding eddies and currents this could be verified with SSH.
L355-358 – The figures showed greater coherence with the mode 2 tides. Would that suggest those are more responsible?
L384 (and in other places) – This study does not directly show that there is mixing due to tides. It is inferred based on surface measurements. This should be explicitly clear somewhere.
L389 – “less turbid”
L395 – Might be helpful to refer to these pathways by longitude here as well
L425 – Run-on sentence. Please reword.
L428 – Do not directly state what your future work is. I think a reworded version of the previous sentence could be used to motivate it without saying it directly.