Intertidal regions regulate seasonal coastal carbonate system dynamics in the East Frisian Wadden Sea

Meyer, Julia; Voynova, Yoana G.; Van Dam, Bryce; Luitjens, Lara; Daehne, Dagmar; Thomas, Helmuth

doi:https://doi.org/10.5194/bg-22-6255-2025

Articles | Volume 22, issue 21

https://doi.org/10.5194/bg-22-6255-2025

© Author(s) 2025. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/bg-22-6255-2025

© Author(s) 2025. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 22, issue 21

Research article

|

30 Oct 2025

Research article |

| 30 Oct 2025

Intertidal regions regulate seasonal coastal carbonate system dynamics in the East Frisian Wadden Sea

Julia Meyer, Yoana G. Voynova, Bryce Van Dam, Lara Luitjens, Dagmar Daehne, and Helmuth Thomas

Download

Final revised paper (published on 30 Oct 2025)
Supplement to the final revised paper
Preprint (discussion started on 23 Oct 2024)

Interactive discussion

Status: closed

RC1:
'Comment on egusphere-2024-3048', Anonymous Referee #1, 02 Dec 2024
Meyer et al. conducted five seasonal cruises in the German Wadden Sea, relying on FerryBox (presumably) surface surveys and discrete samples from its outflow, to assess the controlling factors on total alkalinity (TA) and total dissolved inorganic carbon (DIC). Their results suggest that primary production and benthic contribution both affect water column inorganic carbon system. This study uses standard mixing line approach (for TA and DIC), temperature normalization (for pCO2), as well as reaction stoichiometry in an attempt to tease apart the various processes and changing physical conditions on the carbonate system.
After reading the manuscript, I did not grasp any new information from this study. Several of the coauthors have published numerous articles in this general area with similar conclusions. To improve the writing, I would highly encourage the authors to take a step back and think about how to emphasize the novelty of this work. The writing itself also can use much help to streamline the content, for example the extensive results on turbidity and chlorophyll as well as several spatial maps (for example Figs. 2, 3,7, 12) may be moved to supplementary materials to improve the flow and reduce redundancy, and perhaps merge Figs 4 and 5. Much of the method section can also be simplified because the vast majority of the analytical methods are standard ones hence there is no need to present the procedures in this great length.
Below are some major comments:
First, while the mixing plots are the primary tool being used to separate processes, I find the presentation to be weak. First, Table 1 showed that river DIC endmember for Eastern Wadden Sea in Mar, May and July 2022 had the same value (2647.2 µmok/kg), although this average value was taken from Wester River in Aug and Oct 2021 and Apr 2022. There is no explanation why this was done since the timing doesn’t even match. Given the large, known, seasonal variations (for example Jul and Oct 2021 the values were ~120 µmol/kg apart), this single endmember choice for all three 2022 cruises can easily produce biased results using the following mixing line examination. Aside from the problematic endmember choice, the mixing plots (Figs. 8 and 9) appear to have used the TA and DIC data at both minimum and maximum salinities to draw the lines, instead of the defined endmember values.
Second, even though both TA-DIC and TA vs. DIC plots have been used previously to address reaction stoichiometry, these methods are more of “oceanic” in which relatively narrow salinity ranges are assumed. However, this study covers fairly wide salinity ranges in the coastal sites (for example see Fig. 8) with non-zero and variable river input, I suspect that these approaches are not appropriate in the data interpretation. Some kind of normalization to rid the estuary-coastal mixing needs to fully address the reaction stoichiometry issue.
Third, in separating temperature vs. biological effect on pCO2 variations (note pCO2 does not have “concentration” as several places show), the Takahashi 1993 coefficient applies to the oceanic water under his study, for lower salinity waters with different buffer capacity, 4.2% cannot be directly used.
Below are some minor points that I noted, but certainly this is far from a complete list:
L248, did you correct possible river Ca contribution? See:

Beckwith, S.T., Byrne, R.H. and Hallock, P., 2019. Riverine calcium end-members improve coastal saturation state calculations and reveal regionally variable calcification potential. Frontiers in Marine Science, 6, 169.
L352 paragraph, awkward and unclear sentences. There are many like this in the manuscript. Proper proof reading is perhaps needed.

L364, please show justification why these are considered outliers, the same applies to L370-374.

L403, what’s “real” signal?

L537-539, water with lower Ω has carbonate formation but that with higher Ω has carbonate dissolution? Is this arbitrary for the sake of explaining the data or there is actual proof?

L582, decreasing temperature leads to CO2 (not pCO2) solubility decrease?

L594, what’s another “CO2-consuming process”?

L647 and relevant text in discussion – if carbonate dissolution takes place, how does that help to take up atmospheric CO2?
Citation: https://doi.org/10.5194/egusphere-2024-3048-RC1
- AC1:
  'Reply on RC1', Julia Meyer, 22 Jan 2025
  Dear Reviewer,
  Thank you for your detailed review and constructive feedback. Many important points were raised by the reviewers, and we are confident that we can address all of them, and by doing so significantly improve the manuscript. Please see below our detailed response to the reviewer comments, including explanations.
  Below, we summarize our responses to your major comments:
  DIC endmember: We acknowledge the concern regarding the use of a single DIC endmember for the March, May, and July 2022 cruises. We will clarify in the revised manuscript that this value was based on available Weser River samples due to lack of data for the specific cruise periods. The mixing plots were based on observed TA and DIC values at the lowest salinities, not predefined endmembers. We will reconsider the need for DIC_excess and TA_excess analysis, or the possibility for splitting the study into two papers.
  
  TA-DIC and TA vs. DIC plots: We already addressed that in L348-351. [TA–DIC] cannot be used effectively at low salinities (e.g., < 20) or when [TA–DIC] is < ~50 µmol/kg. In our study, we explicitly excluded data from low salinity areas (i.e., salinities <20) and focused on regions where [TA–DIC] was more applicable for stoichiometric analysis. We present only [TA–DIC] results from the East Frisian WS (Table 3, Fig. 14, Fig. 15) with the same salinities shown in Fig. (8) with salinities above 25. We appreciate the reviewer’s suggestion to consider salinity normalization for addressing estuary-coastal mixing and improving the interpretation of reaction stoichiometry. We would like to clarify that the methodology we employed (Equations 4–7) already incorporates salinity normalization in a more sophisticated form, by accounting for a non-zero freshwater endmember. This approach isolates deviations from conservative mixing and allows for the calculation of ΔTA_p, which was essential to our discussion. While our approach implicitly includes salinity normalization, we agree that explicitly reframing it as such could enhance clarity. Additionally, we acknowledge that further normalization efforts could provide additional insights. In the revised manuscript, we will examine whether explicit normalization of TA and DIC to a reference salinity adds meaningful value to the analysis. If so, we will incorporate these results into our discussion.
  
  Temperature coefficient: We agree that the coefficent can vary with salinity. To avoid generalisation, we will revise the manuscript to provide additional justification for using the 4.2% per degree coefficient. Specifically, we will cite previous studies, such as the work of Joesoef et al. (2015) (https://bg.copernicus.org/articles/12/6085/2015/), which shows that the temperature coefficient shows only small changes over a range of salinities down to about 10-15 (see Table 1). This evidence supports the applicability of the chosen coefficient within the salinity range relevant to our study. Furthermore, we will clarify in the text that the stability of the coefficient is not assumed across all salinities, but rather specific to the conditions observed in our investigation. We thank the reviewer for pointing this out and will update the manuscript accordingly.
  
  Minor comments:
  L248- river Ca contribution: Unfortunately, the limited data for the specific river endmembers during the sampling periods in 2021 and 2022, especially for the rivers discharging into the Eastern Wadden Sea, restricts our ability to accurately represent the freshwater contributions and their variability. The available river measurements were limited to seasonal DIC and TA measurements from the Ems River. This would not cover all relevant rivers in our studied area. We appreciate your suggestion and will certainly consider including flux calcium contributions in future work as more comprehensive data become available.
  
  We will provide clearer justifications for the identified outliers (L364, L370-374), revise unclear sentences (L352) and further clarify the concepts mentioned in L403, L537-539, L582, L594 and L647.
  
  We acknowledge that certain parts of the manuscript require improved clarity and readability. We are conducting a thorough proofreading of the entire manuscript to improve sentence structure, remove redundancy, and streamline the text (including moving some figures to the supplementary material), for better flow. Additionally, we may consider removing the excess DIC_excess and TA_excess data from this study to improve clarity and reduce confusion.
  Thank you again for your valuable feedback.
  Sincerely,
  Julia Meyer
  
  Citation: https://doi.org/10.5194/egusphere-2024-3048-AC1
RC2:
'Comment on egusphere-2024-3048', Anonymous Referee #2, 11 Dec 2024
The article examines the contributions of the Wadden Sea to the North Sea's TA and DIC through data collected from five sampling cruises conducted over a year, providing valuable insights into the carbon budget of this region. While the methods are sound and the interpretations are accurate, the article is challenging to read due to its lack of focus and poor narrative flow.It delves extensively into secondary parameters and marginally relevant epiphenomena, detracting from the central story. Additionally, the discussion section includes a detailed exploration of the use of [TA-DIC] as a proxy for assessing ecosystem metabolic status, which could almost constitute a separate article.The study has significant potential, but the manuscript would benefit greatly from streamlining. It should be organized around a clear central research question, with the text and analyses focused solely on the elements most relevant to addressing this question. To enhance readability and impact, the article should be reduced by approximately 30-50% in length, with large portions of the text and many figures moved to supplementary materials.
Moreover, the extensive focus on the relationships between S-TA, S-DIC, and the influence of salinity on TA and DIC could be simplified by normalizing TA and DIC for salinity, eliminating unnecessary complexity and improving clarity.
Ln 25. It reads that calcification stores CO2, which is incorrect. I m not sure what you call carbonate pump.
repetition, rewrite.

and…and…and, use punctuation.

since the onset/start. I guess we are still in the industrial period

there is more recent meta-analyses than Kroeker et al. on the matter.

I m unsure of what you are calling intertidal region. The intertidal zone is the area free of water at low tide and inundated a high tide. Regions are at wider scale. The Wadden sea is not an intertidal region, it does not emerge at low tide. It is a macrotidal region maybe. I think there is a misuse of “intertidal” throughout the MS.

Besides, a word of description of the intertidal would be good for people non familiar with the Frisian area. Is it a rocky shore, a mud flat (yes), with or without vegetation, seagrass saltmarshes.
In general, I think the second paragraph could be reduced, if not removed as it reads more as a conclusion/outlook material, so that we get faster to the point of the study.
65 make 2 sentences
we already know from previous sentence that we are in the east Wadden sea. rewrite the last sentence more clearly.

Are the cruise 24h day and night, or only day (from what time to what time), how many days? . What time of the day where the sample taken? I guess from reading that there were no nighttime data/samples.

intertidal: Do you mean you sailed at high tide in 2-3 meters of water? or you were cruising in the channels and offshore/ deep subtidal area (I guess).

What scale for the pH? How did you calibrate the sensor? TRIS buffers for SW?

Fig1. Do you have a satellite or bathymetry map instead of a road map? It is difficult to understand the nature of the coastal area.
pCO2obs is the pCO2 measured with the ferry box, or?

So, if I understand well, you applied one end member to one part of your sampling points and another to another half. How did you decide the limit? Doesn’t this have the potential to create an artifact at the frontier between your 2 zones?

155-160. you need to better explain why this equation and what it does.
155-184. These methods need to be better explained. Btw, where do you get the DICocean and TAocean from?
the seasonal highest salinity for Si or Socean, and why the highest, not average or else?

Moreover, if I read well, DICexcess is rather a deficit due to photosynthesis which you attribute to primary production. What if you have a (real) excess, you can’t apply PP stoichiometry as DIC can be emitted by numerous redox reactions from the sediment.
What pH is it? Is it on total scale from TA+DIC, or from the CTD? On which scale?

Why highlighting calcite over aragonite?

Paragraph 280.
Why don’t you directly calculate normalized TA and DIC with your riverine end member with e.g. Friis et al 2003 equation and directly discuss those TAn and DICn variations, due to coastal Production, and drop all the text regarding mixing/dilution/conservative impact of salinity on TA and. Figure 8 could only be 2 plots, one for DIC and one for TA.
the opposite would be quite unusual in a marine system. Unless pCO2 of several thousands. I don’t see what use can be done of the difference between TA and DIC.

415 – 422. The text is full of redundant sentences: March… TA positive… explained by PP etc. June… TA positive… explained by PP. July…
The results section is very long, very (over) descriptive, lack flow/focus/direction and, hence, is tedious to read. Focus on the matter of interest here, the production of TAexcess and DICexcess by the coastal area and move to supplement all description of CHla, turbidity, S-TA and S-DIC relationships, nutrients etc. unless it serves the story you are trying to tell.
this would rather belong to the Material and methods.

Discussion: The lack of focus and the excessive length of the text make the article tedious to read. The [TA-DIC] proxy gains increasing prominence as the article progresses, but it is not introduced in the Materials and Methods section. Instead, it first appears in the results section, leaving readers unsure of its purpose. Only in the discussion section does its relevance and the authors’ intent become clear. This approach makes the manuscript feel disjointed, with section 4.2 reading like a separate article embedded within the main one
The section 4.3 is dispensable, the 4.4 surely as well.
Citation: https://doi.org/10.5194/egusphere-2024-3048-RC2
- AC2:
  'Reply on RC2', Julia Meyer, 22 Jan 2025
  Dear Reviewer,
  Thank you very much for your thoughtful review and constructive feedback. We greatly appreciate the time and effort you dedicated to evaluating our manuscript. Your comments raised many important points, and we are confident that we can address all of them and, in doing so, significantly improve the quality and clarity of the paper. Specifically, we agree that the focus on the relationships between S-TA, S-DIC, and the influence of salinity could be streamlined. We will reconsider the level of detail in this section, reduce redundancy, and concentrate on highlighting the key findings.
  Below, we provide a detailed response to each of your comments:
  Ln25: We agree that the phrasing could lead to confusion, particularly regarding the role of calcification and the carbonate pump in CO₂ dynamics. We will clarify this in the revised manuscript.
  
  To improve readability, we will:
  Reduce repetition and the excessive use of "and."
  
  Clarify the term "onset/start" to refer to the industrial period.
  
  Use more recent citations than Kroeker et al.
  
  consider moving some figures to supplementary materials
  
  Provide a clearer description of the sampling area, including the versatility of the Burchana vessel, with an added satellite map for better context.
  
  65 make 2 sentences: We will split the sentence for improved readability
  
  The cruises were conducted during the day, starting in the morning at low tide. This will be clarified in the manuscript.
  
  The pH calibration method will be added and Fig. 1 will be changed to a satellite map.
  
  pCO₂_obs refers to the measured pCO₂ values obtained from the CONTROS sensor, integrated within the FerrxBox system. We will clarify this in the manuscript.
  
  The boundary between the two zones was determined based on hydrological and geographical characteristics, specifically near the sewer (e.g., Bremen-Farge, Weser). These locations were chosen because they are located at a transition point where land-derived water inputs significantly influence the river's chemical composition. Additionally, we are considering removing the TA_excess and DIC_excess sections and potentially preparing a second publication focused on these data. This would help streamline the current manuscript, making it more concise and focused.
  
  155-160 / 155-184: DIC_ocean was already described in line 159. However, we will revise the manuscript to provide a clearer explanation of the equation. We used DIC_excess as a proxy for primary production but acknowledge other contributing processes such as sedimentary redox reactions. To streamline the manuscript, we are considering removing the DIC_excess and TA_excess analysis and potentially publishing this as a separate study.
  
  The pH was measured as described in the “FerryBox measurements” section. We will add, that the pH values are on the NBS scale.
  
  We calculated also the aragonite saturation. We focus on calcite rather than aragonite because it is more stable in low-salinity estuarine environments like the Ems River Estuary. The system is undersaturated with respect to calcite (Ωcal < 1), meaning that dissolution becomes a more favorable process than precipitation under these conditions. Including calcite provides a more relevant insight into carbonate stability in these conditions.
  
  Normalization of TA and DIC using a riverine endmember approach can be useful, but it also has some disadvantages. It may oversimplify the complexity of carbonate system dynamics by neglecting factors like biological processes, sediment contributions, or spatial and temporal variability, potentially leading to misinterpretation of coastal production effects. Additionally, this approach assumes the riverine endmember is representative across all conditions, which may not always be the case. For this study, we selected endmember salinities based on the observed range in the Wadden Sea to reflect the mixing processes within the study area. Using a zero-salinity endmember—located outside our focus area—could lead to misinterpretation by incorporating processes upstream of the study boundaries and distorting signals of production or consumption. The mixing plots, however, remain valid for distinguishing physical mixing from biological processes, such as photosynthesis, respiration, and CaCO₃ dynamics.
  
  We will clarify this in the revised manuscript and evaluate whether normalizing TA and DIC data adds meaningful insights. If so, these results will be incorporated into our discussion.
  
  We also respectfully refer to our response in Reply to RC1 regarding the reviewer's suggestion on the use of a 'zero' salinity endmember.
  
  We acknowledge the reviewer's concern about redundancy and excessive length in the results section. To address this, we will streamline the text to focus on the core findings. As part of this effort, we are considering removing the DIC_excess and TA_excess analysis and potentially publishing this as a separate study. Additionally, while the TA-DIC proxy is included in the results, we will introduce it earlier in the Materials and Methods section for greater clarity. Sections 4.3 and 4.4 will be condensed or removed to enhance focus and improve the manuscript's flow, while retaining essential information on TA and DIC dynamics. Furthermore, some figures will be merged or moved to the supplementary material to further streamline the main text.
  Thank you again for your valuable feedback.
  
  Sincerely,
  Julia Meyer
  
  Citation: https://doi.org/10.5194/egusphere-2024-3048-AC2

Peer review completion

AR: Author's response | RR: Referee report | ED: Editor decision | EF: Editorial file upload

ED: Reconsider after major revisions (19 Feb 2025) by Frédéric Gazeau

AR by Julia Meyer on behalf of the Authors (03 Apr 2025) Author's response Author's tracked changes Manuscript

ED: Referee Nomination & Report Request started (08 Apr 2025) by Frédéric Gazeau

RR by Anonymous Referee #3 (30 May 2025)

Suggestions for revision or reasons for rejection

Overall reviewer comment:
The authors have conducted a both seasonal and spatial study of the East Frisian Wadden Sea exploring how biogeochemical processes impact the carbonate system dynamics in the area. The study provides valuable insights on the areas importance as a node between land and the North sea. The authors have done a major job streamlining the text since previous reviews, however I believe the manuscript can benefit from a few more revisions before publication. Most importantly, I suggest further clarifying several figures and figure texts to better tie them together with the main body text and reduce potential confusion. I suggest the heavy focus on riverine influence in intro and conclusions be further backed up in the discussion, alternatively increasing the weight on tidal pumping if that is what the authors deem more appropriate. Finally, I suggest the authors simplify the geographical language to increase readability for readers not familiar with the area. See my major comments below for these three points.

Major comments:
Overall comment: The authors use place names extensively throughout the text, example island names. While it might be important to mention I a few cases I suggest thinking about whether this can be skipped most of the time as I don’t think that level of spatial detail is necessary for the broader colclusions the authors want to make. As someone not familiar with the region I found it confusing as well as tedious to have to go back to the main map and check the location constantly. The division of East and West are there, and inside/outside barrier islands could be used for example.

Intro/result: As riverine discharge and land-ocean connectivity is highlighted several times in the intro and briefly in the conclusions as well would have wished for a bit more study site description as well as discussion on riverine discharge. Apart from mentioning a sluice in the results we do not know from the study site description the freshwater dynamics of this region. Are there sluices at more places? How often do they usually open? What about the Weser and Ems river? For example the comment later in the text on the low river flows in summer could be added to the study area description. Adding many mentions of rivers in the beginning of the paper creates an expectation that this will be discussed/important later on. I suggest expanding on this in results/discussion (see example in table 2 comment) or it will feel like this important piece is ignored. Alternatively tone down the weight you put on it in the beginning.

202: The sluice opening being the reason for the salinity drop it seems improbable to me as the salinity anomaly reaches even outside of Nordeney and into Jade bay. I am not an expert in this area so maybe that is indeed the case, however I suggest though thinking about if this could be a larger source of freshwater transported from further away, ex the Weser river or further away.

235/fig 3: I am a bit confused about the conclusion that all points plot above the TA mixing line, to me it is not so clear from the graph. Could this be clarified further in the text? See also suggestions in next comment. To me it is only western May 2022 that plot clearly above a mixing line.

Fig. 3: I suggest explaining/clarifying the lines in the figure as well as the figure text. I assume one is connected to the Eastern WS and one to the Western WS, but it is not mentioned and becomes extra important when the easter/western points are similar. The figure is also not clear on where this is a linear regression or a conservative mixing line. I suggest transferring the same language then to the main body text, eg about “The short mixing line” in line 238.

272: There is a limit of 200 umol/kg [TA-DIC] that is chosen here and used throughout the results and discussion, but to me it is not clear where this value comes from or what it represents entirely. I am not sure if this is the same as the respiration/photosynthesis line mentioned in the sentence after. I propose both these concepts are explained further to situate the reader better.

Fig. 5: I was a bit confused in the main body text (3.4) on where we see these “clear positive trend in dTAexcess”. I suggest you help the reader see and trust the pattern you are referring to better by statistics like a t-test, regression or similar to see if the differences/increase is significant. Additionally, you could display your scatters as box- or violinplots if you want.

Fig. 6: To me as a reader it is hard to verify whether there is a seasonal trend in this figure as we cannot see which points represent march and which represent may. Could this be highlighted by colors like in fig. 5? Or different symbols? I also suggest to make the language around the slopes/lines more clear. The redfield ratio is a fixed ratio of 106:16:15:1 C:N:Si:P. I suggest calling the slopes C:N or C:Si instead to more clearly show it is your calculated ratios. Your main body text about this figure is really clear and easy to understand, I suggest utilizing the same language in the figure text.

405: I think it is really great that you highlight tidal-driven fluxes here as it has been shown to be of importance in this region before (de Groot, T. R. et el. 2023 https://doi.org/10.5194/bg-20-3857-2023, Moore, W.S. et al. 2011https://doi.org/10.1016/j.gca.2011.08.037). As you highlighted that summer river flows are low I think you could emphasise/explore tidal pumping more in your whole text. I think it could elevate your sediment-source conclusions further and strengthen them further by tying them to previous findings in the area.

Minor comments:
Fig. 1: Subsequent spatial maps show more eastward area, like the full Jade bay and Weser river. I suggest staying consistent and keep the same features here in the main map.

2.3 Discrete samples: Can precision of the different parameters/methods be added? And are the CO2SYS constants all here?

168: I would rephrase the wording here. The citations are about calculating conservative mixing between two endmembers, not specifically the river influence.

174: Similar to above. Estuarine inputs are diffuse, please clarify further.

184: Deviations from mixing line is not necessarily only consumption/production, but also addition by sources outside the two endmembers.

142: missing a was after (Ωcal).

239: “(up to 32)”. Could you write this as a range instead? I think it could make it clearer what you mean by higher salinities.

267: I think there is a typo here since there is no fig. 7. Should it be fig. 4 maybe?

275: I might have missed, but the sentence about the salinity change in July 2022 is not entirely clear why it is relevant to TA-DIC here. Could this be clarified in the text?

Fig 4. “The grey dashed line represents the regression slope 280 (- 123 / 138 = - 0.89) proposed by Xue and Cai (2020)”. Could you clarify what this line is “proposed” to show?

286: What is meant by extreme values?

329: “and” missing between dTAp [TA-DIC]?

341: “excess of TA and DIC”. I suggest you clarify what this is an excess from. It has been really nicely explained in previous parts, but out of context from the conservative mixing explanation this could be confusing for a lazy reader.

End of review

Hide

RR by Marta Álvarez (19 Jul 2025)

ED: Publish subject to minor revisions (review by editor) (14 Aug 2025) by Frédéric Gazeau

AR by Julia Meyer on behalf of the Authors (28 Aug 2025) Author's response Author's tracked changes Manuscript

ED: Publish subject to technical corrections (18 Sep 2025) by Frédéric Gazeau

AR by Julia Meyer on behalf of the Authors (24 Sep 2025) Author's response Manuscript

Download

Article (5928 KB)
Full-text XML

Short summary

The study highlights the inter-seasonal variability of the carbonate dynamics of the East Frisian Wadden Sea, the world's largest intertidal area. During spring, increased biological activity leads to lower CO₂ and nitrate levels, while total alkalinity (TA) rises slightly. In summer, TA increases, enhancing the ocean's ability to absorb CO₂. Our research emphasizes the vital role of these intertidal regions in regulating carbon, contributing to a better understanding of carbon storage.