the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Evolution of biogeochemical Properties Inside Poleward Undercurrent Eddies in the Southeast Pacific Ocean
Abstract. Oceanic eddies are ubiquitous features of the circulation through to be involved in transporting water mass properties over long distances from their source region. Among these is a particular type with a core within the thermocline with little signature visible from space. Despite their significance, their role in the ocean circulation remains largely undocumented from observations. This study characterizes the variations in internal biogeochemistry, disparities with external properties, and processes influencing the dissolved oxygen budget of Poleward undercurrent eddies (PUDDIES) during their transit to oceanic waters. Employing a high-resolution coupled simulation of the Southeast Pacific, we scrutinize eddy dynamics and biogeochemical processes associated with the nitrogen cycle, including characteristic mechanisms of Eastern Boundary Upwelling Systems (EBUS) such as denitrification. Our findings reveal that Puddies capture a biogeochemical signal contingent upon their formation location, particularly associated with the core of the Peru-Chile Undercurrent at the core of the Oxygen Minimum Zone (OMZ). While permeability at the periphery facilitates exchange with external waters, thereby modulating the original properties, the core signal retains negative oxygen (O2) anomalies and positive anomalies of other biogeochemical tracers. These disturbances likely contribute to average properties that exceed the 90th percentile threshold in the open ocean, contrasting with the formation zone where they surpass 50th percentile levels. Suboxic cores are prevalent near the coast but decrease in abundance with distance from shore, giving way to a predominance of hypoxic cores, indicative of core ventilation during transit. The principal mechanism governing O2 input into, or output from the eddy core entails lateral and vertical advection, with vertical mixing supplying O2 to a lesser extent. Biological activity consumes O2 for approximately 6 to 12 months more intensely the first 100 days, thereby facilitating the persistence of low O2 conditions and extending the lifetime of biogeochemical anomalies within the core. The ammonium and nitrite depleted out of time in the eddy core with a decay rate greater than the nitrate and nitrous oxide, while these are accumulating in open sea. Our observations suggest that southern regions of the southeast Pacific OMZ undergo greater deoxygenation and nutrient enrichment due to Puddies compared to northern regions. This underscores the significant role of Puddies in modifying biogeochemical conditions in the open ocean and in extending the boundaries of the Southern tip of the OMZ.
- Preprint
(4479 KB) - Metadata XML
-
Supplement
(1161 KB) - BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2024-1290', Anonymous Referee #1, 11 Sep 2024
First, let me apologize for the delay in providing my comment for the manuscript, which greatly affects the review process. Please accept my sincere apology.
In this manuscript, by employing a high-resolution coupled simulation of the Southeast Pacific, Ortiz and colleagues characterized the variabilities in internal biogeochemistry, disparities with external properties, and processes influencing the dissolved oxygen (O2) budget of Poleward Undercurrent Eddies (PUDDIES) and scrutinized eddy dynamics and biogeochemical processes associated with the nitrogen cycle in those PUDDIES during their transit to oceanic waters. Specifically, they documented the evolution of water mass properties and processes inside PUDDIES with contrasting initial characteristics (suboxic versus hypoxic) in order to evaluate the role of these properties in the maintenance of oxygen minimum zones (OMZs). They found that Puddies capture a biogeochemical signal contingent upon their formation location, particularly associated with the core of the Peru-Chile Undercurrent at the core of the OMZs. Furthermore, they found that while permeability at the periphery facilitates exchange with external waters, the core signal retains negative O2 anomalies and positive anomalies of other biogeochemical tracers. These disturbances contribute significantly to average properties that exceed background conditions in the open ocean. The main process controlling O2 input into, or output from the eddy core entails lateral and vertical advection, with vertical mixing supplying O2 to a lesser degree. Biological activity consumes O2 inside PUDDIES for around 6 to 12 months, especially intensely for the first 100 days, thus facilitating the persistence of low O2 conditions and extending the lifetime of biogeochemical anomalies within the core. The authors suggested that by representing more complex biogeochemical and physical processes in ocean models, future studies will be able to quantify the effects of other SMS processes within Puddies that are not yet considered in their study. Validation and supplementary statistics of Puddies through further observations are urgently needed.
The manuscript is well written and the method and model setup are well-designed. The authors also did a thorough job on the transformations and transport of biogeochemical elements inside PUDDIES. I have learned a lot and have enjoyed reading the manuscript. My only reservation regarding the manuscript is that I feel the authors can do a better job on producing Figures. Some figures (especially figure 1) are small with a bad color scheme that makes it difficult for readers to see and interpret the (sometimes) dense information that the authors try to convey.
Other detailed comments.
Line 17: Please change to: “Oceanic eddies are ubiquitous features of the circulation thought to be involved…”
Lines 30 - 31: “These disturbances likely contribute to average properties that exceed the 90th percentile threshold in the open ocean, contrasting with the formation zone where they surpass 50th percentile levels.” Please consider adding a clear definition of the percentile threshold levels here.
Line 37: What do you mean by “out of time”?
Figure 1: Please consider improving the quality of Figure 1: Bigger size, bigger font for texts, different color scale other than the rainbow, clearly distinguish between black lines and black contour, clearly show the yellow dots tracking the Puddies… They are all very hard to see now.
Line 189-190: Please change to “which is a scalar”
Line 201: Is the Score defined as density core? Please clarify (I know you defined it later but it should be defined clearly earlier).
Line 237: Please change to N2O production
Line 369-370: Please change “The correlation coefficients varied between 112.5 and 457.45” to “The slope values…”
Line 490 and Line 493: please add < 0 or > 0 delta O2/delta z here.
Table 6: Probably a naive question, and please excuse my ignorance, but can you clarify what are x and y used here to calculate the decay rate k?
Line 614: Please change to: “affected also by the southward reduction in contribution of ESSW”
Line 624: Please replace Table XX here with the correct table.
Line 690: This is the first instance in the text that I see the acronym CARS. Please define it.
Citation: https://doi.org/10.5194/egusphere-2024-1290-RC1 -
RC2: 'Comment on egusphere-2024-1290', Anonymous Referee #2, 01 Nov 2024
Review of Ortiz et al. "Evolution of biogeochemical Properties Inside Poleward Undercurrent Eddies in the Southeast Pacific Ocean"
The paper by Ortiz et al. aims to characterize the biogeochemical evolution of poleward undercurrent eddies (PUDDIES) generated in the Southeast Pacific OMZ. To do so, the authors use 9 years (2000-2008) of output of a physical-biogeochemical coupled model (ROMS+BioEBUS) run on mesoscale resolving grid (1/12 degree resolution ~8km) and saved as 3-day means. The authors distinguish between the evolution of PUDDIES characterized by initial hypoxic conditions and those with initial suboxic conditions, contrast the evolution of water mass properties inside and outside of the eddies and focus on a few case studies to determine changes in the eddy properties along their trajectories. The authors use Reynolds decomposition and eddy tracking methods to study the impact of the PUDDIES on the oxygen and nutrient fields.
General comment
The topic and idea of the paper are interesting, but I’m not convinced by many of the choices made by the authors and I find the paper difficult to read and its results difficult to interpret. I have decided not to go into the details of the results for this reason, but to comment on the methods at this stage. I think this manuscript and its related research needs major revisions before being published. I’ll be happy to provide a more detailed review on the results once the methods are better justified.
Many of the choices in the methods need to be clarified. The domains in which the region of study is split (8 domains) seem odd and not well justified in the methods (especially the latitudinal boundary, which seems arbitrary). Some of these domains are very narrow and coastal (H) where the authors state the algorithm has trouble identifying eddies (lines 299-307), some mix coastal and offshore zones (G,F), some are very wide (E). All have irregular shape and mix zones at a variety distances from the coast (where eddies are generated). This makes for a statistical and geographical nightmare in understanding and interpreting the data. This is a major problem for me, because all the results are presented in terms of averages across these oddly shaped and dishomogeneous regions.
The authors interpolate 37 vertical layers of model output to a regular grid of 160 vertical layers generating a lot of “fake data” for their analysis. The choice of “reference average” for the Reynolds decomposition is unclear. The authors say that they use a circular eddy mask to extract the eddy properties, when the Faghmous algorithm provides the actual shape of the eddy contous, which provides a more accurate definition of the eddy core properties. A lot of these choices need to be better justified and a few probably should be revised.
The authors also use 9 years of model output at 3-day mean resolution, but then only use two eddy tracks (among likely tens or maybe hundreds found by the employed algorithm) as “case studies”. There is no proof that the two tracks are representative of the eddy populations. Why not building an average of the eddies in time along their tracks so that eddies of the same age are averaged together? This was successfully done in previous studies (https://doi.org/10.5194/os-15-1111-2019 ) and it would allow to use a statistical approach that really takes advantage of the large amount of information provided by the model data. The model data available for the study is so abundant that the current approach seems reductive.
I have brought up many of these concerns also in the detailed comments.
The paper is also difficult to read due to the Methods section being poorly organised and missing important detail and the figures not having appropriate labelling, so the reader has to note what the figures represent by hand (impossible to do if reading the paper online). The Supplement is currently unreadable, see more comment below.
Detailed comments
1. Line 49: “Under these conditions, heterotrophic metabolic processes prevail” - what conditions? Prevalence of heterotrophic processes is mostly due to being at depth and hence below the sun-lit layer.
2. Line 60: “where anoxic conditions can even be observed” - this piece of sentence need revising, it doesn’t read correctly in English
3. Line 67: “through the global warming” - please remove “the”
4. Line 72: “turbulent dynamics” - I’d be careful here using the word “turbulent” since turbulence is a very broad term that in physical oceanography describes a large variety of scales of physical processes down to sub-meter scales. I would rather use “meso- and submesoscale” instead, which is more in line with the topic of the paper. Some processes such as “turbulent mixing” are most likely parameterised in models, but this isn’t what the authors refer to.
5. Lines 72-55: I would add some support to the claim of the importance of mesoscales in the OMZ representation in models, currently unsupported in the paper. Literature suggestion: https://doi.org/10.1029/2022MS003158
6. Lines 117-118: Unclear sentence
7. Line 196: What about zooplankton respiration?
8. Line 203: The authors interpolate the model output to a regular grid of 5m of vertical spacing between 0-800 m depth. This means interpolating to 160 vertical levels! The model output has only 37 vertical levels, of which 13 are found “in the deep ocean”, hence possibly only 24 model grid levels are in the range of interest. Why generating so much artificial data?
9. Line 204: “In the deep ocean (~4000 m depth) typically 13 of the 37 vertical levels fall within this depth range.” - You need to indicate the specific depth range, ~4000 m is not a range.
10. Lines 207-212: I don’t understand the need to split the domain at 30°S. Is there a reason why this exact latitude is significant?
11. Line 211: “Pudies” - Puddies? Also, the choice of acronym must be homogenized across the manuscript: all capital (PUDDIES) as in the abstract, or capital P only (Puddies) as here?
12. Lines 213-214 and Figure 1c: If the first 100 km are such a relevant range for the formation of Puddies, why not looking at zones that have boundaries at equal distance from the coast, so that the 0-100 km zone is the formation zone across the whole system and then more offshore zones can be defined?
13. Lines 217-228: What is your definition of mean state? This must be made explicit here. Is it an overall mean across your 9 years of simulation? In this case, seasonal variability will contribute to the “fluctuations” as it will be part of the residual field. Or else, is it something like interpolated seasonal or monthly climatological means? Please specify.
14. Line 261: Did the algorithm identify eddies larger than 300 km of diameter, and if so how many? This seems very large and surprising to be identified as eddy for the Faghmous et al. 2015 algorithm run on an 8km grid.
15. Line 262: “every three days” - does this simply refer to the fact that the model output was saved as 3-day means? If so, please rephrase, since a 3-day mean is different from “every three days” output, which might mean a simple snapshot of the model every three days.
16. Lines 247-264: How many eddies were identified in total by the algorithm over the model output? How many independent tracks did they correspond to? This is important information to be included here.
17. Lines 265-271: I’m puzzled by this choice of only analyzing the tracks of only 2 Puddies when the authors have 9 years of model output available at three days mean output resolution, which most likely allows the authors to analyze tens or maybe hundreds of eddy tracks. Is there any reason why these two tracks should be particularly relevant or representative of the mean eddy for the northern and southern subregions?
18. Lines 286-290: Why do you apply a circular mask that will surely miss part of the eddy shape and likely include part of non-eddy waters (eddies are rarely perfect circles), when the Faghmous et al. 2015 algorithm already provides the identified eddy perimeter mask?
19. Line 300 – how do you define a “puddy profile”?
20. Lines 304-307: This seems like an important caveat and it needs better clarification. What does “slightly overestimated” mean in numbers? Please provide a number of how many identified Puddies are actually not Puddies but coastal trapped waves or other structures. In other studies, coastal eddies in the first life stages have been excluded from the analysis just because the Faghmous et al. 2015 algorithm was having difficulty identifying them.
21. Figure 2: Please add labels to this figure, what does each line represent? It’s not enough to have it in the caption, it makes the figure really difficult to understand.
22. Table 1: Number of pixels? What does it mean?
23. Tables (in general), Results and Methods: I’m really not convinced by having these 8 different regions. It makes the results overly complicated to understand. I think the authors should rethink this analysis almost entirely and use more regular domains, for example defined by offshore distance bands.
24. Tables (in general): These tables are heavy to read. They need to be summarized into figures, bar plots, or something that makes it possible to the reader to grasp the results.
The supplement is currently unreadable. It’s impossible to revise a document where figures and captions are provided separately in different PDFs. I had to copy by hand the text onto a word document and then screenshot the figures and stich them on top of the correct captions in the word document, to be able to understand what I was looking at. The quality of the supplement should be at least checked by the journal upon submission. The supplement should be revised entirely to make it clear to the reader. There are also typos in the supplement’s captions and figure axes miss titles and key information.
Citation: https://doi.org/10.5194/egusphere-2024-1290-RC2 -
EC1: 'Peer-review report #3 (EGUSPHERE-2024-1290)', Olivier Sulpis, 07 Nov 2024
Dear authors,
Please find below a third review for your manuscript. This review arrived after the discussion period was closed, so I am posting it here. Please reply to all three reviewers in your final response.
Best regards
Olivier Sulpis
_________________
Review of: “Evolution of biogeochemical Properties Inside Poleward Undercurrent Eddies in the Southeast Pacific Ocean” by Ortiz et al.
The manuscript combines a comprehensive observational dataset and physical-biogeochemical numerical simulations to analyze the origin of Poleward Undercurrent Eddies (Puddies) and their transports of biogeochemical properties in the Southeast Pacific Ocean. The topic is interesting and relevant. The manuscript is well structured and presents enlightening results about the subsurface mesoscale structures formed in the Peru-Chile Upwelling System. In my opinion, the manuscript can be published after addressing the minor comments listed below.
MAJOR COMMENT
- Some of the most important results are based on the definition of the isopycnal surfaces that bound the OMZ. This should be justified. Although they can be supported by the literature, these values might not be adequate for your model data.
MINOR COMMENTS
- Pg1, L22: PUDDIES, Puddies, use only one convention.
- Pg4, L119: puddies, Puddies.
- Pg4, L134: Mention the resolution here (by the way, that is not that high).
- Pg4, L138: were, was.
- Pg4, L144-145: “…resolution of… 37 levels… suitable for resolving mesoscale features”, please discuss briefly about this.
- Pg4, L147: 20°, 20°S.
- Figure 1: Dots of the trajectories of the Puddies are too small.
- Pg5, L154: “Black lines…”, where? In panel (b)?
- Pg5, L157: “Yellow dots”, where?
- Pg5, L160: “the possibility that the same eddy…”, has you considered using an eddy-tracking method to avoid accounting for the same eddies at the same grid cell?
- Pg5, L178-179: “Phytoplankton biomass was estimated…”, what about the one from the model?
- Pg6, L195-196: “For O2, the source…”, any reference?
- Pg6, L199: “…is centered at a density surface of …”, how is this assumption supported?
- Pg6, L202: “…the 1000 m and 500 m depth isobaths”, where?
- Pg6, L211: Pudies, Puddies.
- Pg7, L225: “…with annual and interannual variability”, and seasonal?
- Pg7, L229: Define AOU, <Delta>NO3, <Delta>N2O.
- Pg7, L235: Substitute “Eq. (3)” by “relation” or “equation”.
- Pg7, L236: Explain what subscripts “sat” and “obs” mean.
- Pg7, L237: Sarmiento, Sarmiento’s.
- Pg7, L238: Delete “Eq. (4)”.
- Pg7, L240: Delete “Eq. (5)”.
- Pg8, L253: How are “S_upper” and “S_lower” defined?
- Pg8, L254-258: A scheme would be useful.
- Pg8, L266-270: What about uncertainty?
- Pg9, L286: Can a polygon be circular?
- Pg9, L304-306: What about a second criterion (for example, Okubo-Weiss parameter or so) to better define the eddies?
- Pg9, L305: Eddies? Eddy signals? Eddy-like features? Define what you mean exactly.
- Pg11, L334: Puddies profiles, Puddie profiles.
- Pg11, L335: The number, The numbers.
- Table 1: Depths are usually defined as positive, in contrast with the vertical coordinate. The total area (and not just the number of pixels) can be interesting.
- Pg12, L352: between, among.
- Table 3: Define “R”.
- Figure 3: You should move the labels “(a), (b),…” out of the small boxes, and also make them bigger.
- Pg14, L382: Corresponding, correspond.
- Pg14, L384-385: “Vertical error bars… of O2”, unnecessary. It could be something like: Error bars correspond to the standard deviations.
- Pg15, L409: “()”, underbars?
- Pg15, L410: “()’”, primes?
- Pg16, L429-431: “…the variability of… with oxygenation”, dispersion of the eddy trajectories?
- Pg17, L436-437: “Vertical error bars… of O2”, unnecessary. It could be something like: Error bars correspond to the standard deviations.
- Pg18, L451: “denoted by PHYS”, where? Do you mean “referred to as”?
- Pg19, L508: positive vorticity, anticyclonic?
- Pg 21, L544: puddy, Puddy.
- Pg21, L550: positive vorticity, anticyclonic?
- Figure 7: Labels of the color bars are too small.
- Figure 8: Add numbers in the horizontal axes at the two upper rows. Background grids in the plots would be useful.
- Pg23, L574: puddies, Puddies.
- Pg24, L589: Delete “on”.
- Pg25, L596: Insert comma after “waters”.
- Pg26, L665: “within eddies”, surface eddies?
- Pg27, L681: due additional, due to additional.
- Pg27, L682: by-products, byproducts.
- Pg27, L697: Mention the resolution.
- Figures A7 and A8: The points of the results are too small, use different colors between each other for more clarity. Avoid overlapping of the points’ labels.
Citation: https://doi.org/10.5194/egusphere-2024-1290-EC1 -
AC1: 'Reply on EC1', Lenna Ortiz, 11 Nov 2024
Dear Handling editor,
I appreciate your response and the reviewers, we are working on their observations.
We will send the response as soon as possible.
Best regards,
The authors
Citation: https://doi.org/10.5194/egusphere-2024-1290-AC1
-
AC1: 'Reply on EC1', Lenna Ortiz, 11 Nov 2024
Viewed
HTML | XML | Total | Supplement | BibTeX | EndNote | |
---|---|---|---|---|---|---|
275 | 76 | 23 | 374 | 23 | 12 | 13 |
- HTML: 275
- PDF: 76
- XML: 23
- Total: 374
- Supplement: 23
- BibTeX: 12
- EndNote: 13
Viewed (geographical distribution)
Country | # | Views | % |
---|
Total: | 0 |
HTML: | 0 |
PDF: | 0 |
XML: | 0 |
- 1