Review: Spatio-temporal variability of greenhouse gas concentrations and fluxes in shallow coastal bays of the western Baltic Sea by Zinke et al.

General comments

The study investigates an important topic, GHG emissions and dynamics on shallow coastal areas. Every single study is important in gathering more experimental data and it is not necessary to have all aspects addresses in single study. I recommend publishing this study after addressing a few minor issues in the final version of the manuscript.

Introduction

The introduction is fluently written and covers the relevant literature.

Observations

The observations are done using suitable instrumentation. The study does not describe the calibration methods for the gas analysers or whether standard gases were measured during the experiments as a reference. Could be beneficial to add few words about the potential field calibrations.

Results

The study provides a short glance on GHG dynamics on a few coastal bays in Sweden, with high variability observed. The observations were done always around the noon. While this ensures possibility to compare the results between the locations, the study cannot provide any information on diurnal variation and impact of meteorological conditions (e.g. solar radiation)  influencing significantly the biogeochemistry and sea-atmosphere fluxes, especially during the biologically active period. Please address these issues when generalizing the results. The study covers only two seasons, spring and autumn – what is the expected seasonal behavior of concentrations and fluxes?

pCO2, pCH4 and pN2O has been measured for long period throughout the Baltic Sea on ICOS VOS line between Travemünde and Helsinki. Please compare the coastal concentrations with concentrations observed over the open sea and give a rough estimate how many percent of the shallow and bay areas cover the Baltic Sea. This will give an indication of the impact of shallow areas on total GHG budget of the Baltic Sea.

Specific comments

Table 2

Östra Lermaren, please check the values for pCO2 and correct.

Please re-check the pCO2 saturation calculations

There are a couple of typos (comma vs dot), please correct.

How sensitive are the results for the choice of selected wind speed (2 m/s) used in the study? Please calculate the fluxes using wind speeds of 1 m/s and 5 m/s and describe the impact of choosing 2 m/s on the results.

In shallow waters, ebullition may be important for CH4 fluxes – please shortly address its potential importance on calculated methane fluxes in the paper.

Data availability:  EGU journals require open data policy.  As the most important research contribution of this study is the dataset collected,  please submit the data to MEMENTO and SOCAT databases after the publication of the paper so that they will be available for the Baltic Sea GHG research community in the future.

Review: Spatio-temporal variability of greenhouse gas concentrations and fluxes in shallow coastal bays of the western Baltic Sea by Zinke et al., https://doi.org/10.5194/egusphere-2025-4446

The manuscript by Zinke et al. presents data of the concentrations (and fluxes) of the three major greenhouse gases CO2, CH4, and N2O in a variety of shallow bays in the wider Stockholm Archipelago, Baltic Sea, using continuous recording of a system with an air-sea equilibration system coupled to a Picarro G 2508 CEAS. Measurements were done during two seasons, around midday, and a variety of potentially controlling environmental parameters was recorded as well. The authors use these data to calculate fluxes using the ASE parameterization developed by Cole and Caracoa (1998) for lake environments.

While the data set is interesting and the question of GHG emissions from shallow coastal waters and its potential changes is timely, the paper unfortunately does not explain parts of the methods and approach, and despite the fact that the authors claim the need to find ways to address the large spatiotemporal variability of GHG fluxes from these environments, they do by far not fully exploit their data set to answer these questions by e.g. stringent correlation analysis. The only correlation they address in more detail is the one between N2O and CH4, while the physical drivers including the -nicely introduced – topographic openness index – are not addressed in a systematic way.

Therefore, I can recommend publication of the paper only after major revisions.

I will start with some general comments and then get more detailed .

Scope: in the last paragraph of the introduction, the authors indicate the scope of their study, in particular naming the spatio-temporal variability and the examination of potential control parameters. However, they do neither discuss the limits of their approach towards these goals nor fully exploit them.

Variability: it is not really clear how and when the sampling was done exactly. The authors state that the measurements were done during midday (line 81), but also describe a 45min measurement cycle which was repeated several times. From the details of the measurements mapped in the Figs Appendix A1-A6, it appears that for the inner bays, only the long-term measurement at the positions indicated by red triangles were used. For “outside Bay area”, there is no stationary point indicated so apparently the data outside a line indicating the boundary between inner- and outside bay were pooled. The approach of these outside Bay area measurements is not explained at all in the text, nor the selection of the separation or the question how “out” these areas actually are (in terms of connection to the open Baltic, residence time, depth etc.). These approaches should be explained in more detail.

Temporal variability: while it is correct that the instrumentation used by the authors can be used to tackle temporal variability, it is not used that way in this study. Basically, the authors claim to have done experiments over midday, and it is not clear what time frame is represented in their individual data sets (as shown in the Figures A1-A6 of the appendix; a few hours is my guess). There is a complete lack of discussion on diurnal cycling and potential bias, in particular on the fluxes, while very relevant studies on the topic exist, e.g. the study by Honkonen et al. 2021 on diurnal variability of pCO2 fluxes at Utö (really nearby) or the recent study of Pönisch et al. (2025) on summerly GHG fluxes from a rewetted peatland (shallow coastal water). The latter has a detailed analysis on diurnal variation and the effect and biases this might have on GHG flux calculations, which the authors should address.

Flux calculations: The authors should consider whether the flux calculations should be part of the paper. The strength of the manuscript is the measurement of concentrations and potential relation to controlling parameters. The flux calculations are based on the concentrations measured, a chosen ASE model (without discussion of the choice), and with a lack of wind data. If the flux part should be incorporated, it would be essential to

• discuss the potential daily bias in these shallow systems (see above).
• to at least get access of wind data from an adjacent wind station of a reliable wind product. Just assuming an average wind speed of 2m per second is not state of the art.
• To discuss the chosen wind model. This is of particular importance, as the wind model used in this study has a considerable wind-independent additive term, which dominates at windspeeds < 3.5 m/s. For systems like the ones investigated, there is no right or wrong choice, but some justification would need to be given. Way more important is that it would be important to look into the wind parameterizations used in the flux estimates worldwide the authors compare their results with (Table 5), which likely mostly used parameterizations without such a strong low-wind component.

Assessment of drivers and potential controls:

The authors nicely introduced potential drivers of GHG production in the intro and beginning of the method, and took measures to investigate those (T,S, topographical openness, organic matter content, nutrient concentrations, …). However they report correlations only sporadically and without a sound estimate of significance etc. This is already documented in the separation of the GHG data and the “auxiliary” data in tables 2-4. The authors should try to give statistical underpinned information on correlations and potential driver analysis. A very good example for this is the use of Spearman correlation coefficients and effect size attribution as e.g. in Pönisch 2025.

The only quantitative analysis of correlations in this text is the one between N2O and CH4. While some of the discussion point in this section are really interesting, the interpretation of the correlation (and of the discussion on CH4 and N2O concentration suffers from the beforementioned lack of detailed assessment of the physical drivers. Here, for instant, it is clear that the main control of N2O is temperature, and this is likely also true (mentioned in the text) for methane, in the opposite way, as in the one case, control on solubility is dominant (N2O), while in the other, the control on production seems more important (CH4).

Stringent analysis could potentially help to reveal why and when the inverse correlation of CH4/N2O changes the direction at CH4 higher 250 nM.

As pointed out correctly by the authors, the high variability in shallow coastal waters need more data, but also the best possible research to assign drivers and controlling parameters. Here, the paper could be largely improved.

------------------------

Some areas of concern:

Askö Data: In Section 3.1.4, for the N2O -CH4 relationship, the data from observations in Asko are suddenly introduced, completely “out of the hat”. Asko is a different setting, the method, sampling etc. has not been introduced, and there is no real link to the rest of the study. It honours the authors that they want to publish the data set, but it does not connect to the rest of the study. So I strongly recommend to remove this part, which is not introduced in the methods or site description, nor attached to the study. If these data are “left overs” they can be easily uploaded with relevant metadata to a publicly available data base. It appears that the authors were a little unsure here themselves, as if I am not mistaken, the data are not plotted correctly (Fig 3a). I am very sure that the data referenced as March data are actually September data and vice versa.

Method information:

There is a lack of information on some of the methodological aspects: how and when was the sensor calibrated with which kind of calibration gases? What is the expected accuracy / precision of the measurements? Which data were used for the calculation of the concentrations of the “inner bay” and the outside Bay (see further above). When calculating mean values (e.g. for fluxes), were they averaged simply over the No of regions, or were they area-weighted. Please add the information needed to assess the methodological part of the manuscript.

Missing N2O data for Bodviken: this should be explained. It is mentioned several times in the text, and difficult to understand, as the authors use an instrument measuring all 3 gases simultaneously, and usually the instrument fails completely or not; there is surely an explanation, but please explain in the text.

Minor issues:

Abstract:

Line 7: “seasonal variation concentrations” – Wording

Line 10: It is unclear why the finding of a shift in N2O-CH4 relation slope indicates a shift in biogeochemical processes; see also comments further above

Line 11: what is meant by “anthropogenically degraded”?

Line 12:”methane emissions that surpasses CO2 update”: first of all, most regions had no CO2 update; second: this is not clear without introducing the fact that the authors calculated CO2 equivalent fluxes – WORDING

Intro:

Line 33: strictly speaking, N2O is a by-product of nitrification, but an intermediate of denitrification

Line 37-38: statement on aerobic oxidation needs a reference

Line 39-44: the manuscript completely ignores anaerobic methane oxidation, both here, and also in discussing the effect of T on methanogenesis; in fact, both methanogenesis and methanotrophy are T-dependent.

Line 55-64: see comment above; the authors did not really fully exploit the possibility to scaling based on driver analysis, nor did they really make use of the high temporal resolution in this study.

Methods:

Chapters 2.1 and 2.2.: as stated above, give more details on method, and introduce the outer Bay sites; also please explain which data were used for the inner and outer Bay mean concentration value (just on red triangle spots, or all data in the inner Bay; outer Bay I cannot tell, it is so far not indicated in the text at all. Also, be more specific about the timing and duration of the data acquisition (midday vs. several hours of measurements).

Line 95: Partial pressures do NEVER have the unit ppm

Line 98: “ atm=106 ppm” ; same, please be scientifically correct here

Section 2.2.3. See remarks above on wind speed data availability, discussion of use of ASE exchange parameterization and its implication for comparison of fluxes with other studies

Section 2.3.4: Line 158: “..most relevant for …” - not fully true for methane, where deeper sediment information might me needed to understand flux or ebullition behavior

Results and Discussion

3.1 Outside Bay areas not mentioned or described before or anywhere after in the text (se general comment further up)

Lines 171-173: The lack of consistent pattern between bay openness and CO2 concentrations (see more general comment above on quantitative correlation analysis) indicates that there is no major control of this parameter on CO2 concentration, but not that there are high spatial and temporal variability in CO2dynamics”. This is no substantial statement.

Lines 184 and 185 „... share ...GHG emissions“ ; Bold and unsupported statement, based just on the similarity of a mean concentration value (for CO2). So why is this used to speculate on GHGs in general. For CH4, for instance, it is well established that freshwater and brackish wate rare quite distinct due to the role of processes involving sulphate.

Lines 188-190. This might be a good place to extend on related work who actually did that, also showing how important it is, like e.g. Honkanenen et al. 2021 or Pönisch et al. 2025. However, this is again a statement true for all three GHGs, so maybe the best place to discuss this, and with it the limits of the study here, would be after the reports on the individual GHGs.

Line 204-206: „while ...aerobic methanotrophs: Missing reference. Also, there is also a wealth of literature on rooted vegetation actively transporting methane, which would escape the flux measurements. This should be briefly mentioned as well.

Lines 216-217: while it is true that the system would measure dissolved methane from the seafloor and dissolved by bubble dissolution, it would be then important to state that bubble-mediated transport itself escapes the device, and in fact that trapped bubbles could be an issue for the measurements. Maybe the authors did some work to protect the inlet from ascending bubbles? If so, it would be good to mention this in the method section.

Line 210: enhanced temperatures also enhance aerobic oxidation rates, though apparently the effect on methanogenesis “wins” here.

Line 211-212: Stratification would enhance concentrations below the stratification gradient, but lead to lower concentrations in the top layer. Please be clear in your argument here.

Lines 216-217: while it is true that the system would measure dissolved methane from the seafloor and dissolved by bubble dissolution, it would be then important to state that bubble-mediated transport itself escapes the device, and in fact that trapped bubbles could be an issue for the measurements. Maybe the authors did some work to protect the inlet from ascending bubbles? If so, it would be good to mention this in the method section.

Chapter 3.1.3 as mentioned in general comments; a quantitative discussion of N2O concentrations in relation to temperature is needed, which is clearly the dominant driver. Or the discussion could be based on the calculated N2O saturations which eliminate this effect. Here, it would help to distinguish whether a part of the difference is temperature rather than production related.

Tables 2-4: again consider the general comment on integrated research on the drivers and the concentrations.

Chapter 3.1.4. while some aspects of the discussion are quite interesting, the analysis suffers from the investigation of T as mean driver before; here again having a look on whether deviations could be found on the saturation level could overcome this. Most of the N2O data will be on a straight line then, and it would be interesting to see how the trend at CH4 > 250 nmol displays on the saturation level.

Still, see my general comment on semi-quantitative statistical evaluation.

Line 268-272: “Since …. . … anomalous patterns”. This section is purely speculative.

Figure 3: Plot a should be removed, as not part of the presented study (and likely wrong seasonal attribution)

3.2.

Line 283-284 „... similar to the complex ...“ This similarity is an artifact of using basically a mostly constant transfer coefficient (i.e. fixing the wind speed); see general comment.

295-296: Again: if flux discussion remains, nearby wind speed should be addressed, as well as the model of use and its impact for the flux comparison of table 5.

3.3.

Line 301 “ … with a median …” Here and elsewhere: please describe how data were averaged (just one No pe area, or “area weighted …”.

Line 309: To assess the regional significance, one would need to know typical flux estimates from similar areas on land or over open water.

Figure 4: I found it really interesting that Högklyeviken is characterized by highest CH4 fluxes, but also as the only region with a net carbon dioxide uptake in both seasons; do the authors see any reason for that (I cannot see a link to the fact that the phosphate-binding experiment took place there …). This is just reviewer’s curiosity, so ignore if not worthwhile pursuing …

Conclusion

Line315: „.. shallow Baltic Sea“ – I think the authors should make clear that the data are from a very small part of the Baltic Sea and surely not fully representative for the (entire) Baltic Sea.

Line320 : …dominates CO2-equivalent fluxes.

Line 324-329 surely need revision after addressing the review’s comments.

Line 332: “However …. budgets”  - shallow bays are not really overlooked, and so far, the data presented were quite in line with earlier findings; so I do not understand the statement

Line 347: Data Availability:

It is of utmost importance that all data , also the auxiliary data, are publicly available upon publication of the paper, once accepted.

Review: Spatio-temporal variability of greenhouse gas concentrations and fluxes in shallow coastal bays of the western Baltic Sea

Zinke et al. (2025)

The manuscript by Zinke et al. investigates the concentration distribution (and fluxes) of the greenhouse gases CH₄, CO₂, and N₂O in surface water in various bays of the Stockholm Archipelago in two seasonal campaigns (spring and fall). The aim was to characterize the spatial and temporal variability of greenhouse gas concentrations (and fluxes) in the surface water of these underrepresented systems and to identify important environmental drivers. The authors recognize, above all, a pronounced variability in the concentration distribution of greenhouse gases.

General comment

The manuscript deals with an important topic, as the significance of shallow water areas as sources or sinks of greenhouse gases is not yet sufficiently understood. This is mainly due to the insufficient data collection in these areas to date. The manuscript makes a contribution in this regard, which is also its strength. However, the manuscript provides little insight into the environmental drivers that influence the source/sink strength of these areas (in this case, bays). This may be due, among other things, to the fact that the results are not clearly discussed and correlations have not been clearly identified. The authors refer too often to the appendix instead of addressing this central point in the main text. In the main text, the reader is confronted with overloaded tables and is expected to work out the connections for themselves. This was not always possible for me, and often the connections suggested in the text contradict the data shown in the tables or Figure 2.

My opinion is that the manuscript needs extensive revision before it can be proposed for publication. The discussion should be more detailed and, above all, the correlations between greenhouse gas concentrations and environmental factors/drivers should be better elaborated (the main objective of the manuscript, as stated by the authors). The presentation of the results should also be revised to make them more accessible to the reader.

Specific comments

Abstract

Line 5    Why was the highly productive summer omitted? What was the strategy here in selecting the study periods?

L6          “relate to bay characteristics and seawater properties”, What does that mean? Express it better and define it more precisely.

L9          “indicating a shift in biogeochemical processes”, It is very unclear what is meant here. This needs to be explained better here and also addressed more thoroughly in the discussion later.

L11        What is meant by “anthropogenic degradation”? This needs to be expressed more clearly.

L13ff     At the end of the abstract, the uniqueness/novelty of the manuscript should be emphasized more clearly.

Introduction

Overall, the introduction is well written. However, it is noticeable that in many places the references supporting the statements have not been cited.

L28ff     References are missing

L31ff     References are missing

L36        Not only refer to your own publication from the group here, but also name the original publications (at least one by another author).

L38        References are missing

L40        “Methane” or “CH4”, This must be consistent throughout the text.

              What does the author mean by “delicate equilibrium between methane production and oxidation”? This needs to be explained better.

L42        References are missing

L44        Here, the sentence needs to explain more clearly how “promotes organic matter accumulation” affects the O2 content in the water.

L50        Reference to eutrophication in coastal ecosystems is missing

L52        CRDS technology can no longer be promoted as a novelty

L55        Discuss here why these periods were selected > Sampling strategy

L55ff     This is where the manuscript's main weakness lies: how do which drivers affect GHG concentrations? This seems to be one of the main objectives of the manuscript.

L57ff     Present more clearly in the discussion how the environmental factors mentioned affect GHG concentration/emissions: eutrophication gradient, geomorphology, physical factors (water retention time), sediment composition. Can the manuscript provide a clear answer to this complex question?

L59ff     “show distinct spatial patterns, with hotspots emerging in different niches within a bay”, where in the following discussion are spatial patterns and hotspots addressed? Should this be evident from the spatial surface data in the appendix? If so, it needs to be given more weight and better presented in the main text.

Methods

L66        What does “continuous daytime measurements” mean? During what period (times of day) were the tests carried out? It should also be mentioned here at the beginning when exactly the investigations in the bays took place.

L73ff     Here, the reader is confronted with a multitude of references intended to show that Ea is a measure that affects GHG dynamics in bays: “all of which can influence GHG cycling.” Ultimately, the authors must ask themselves whether their data really demonstrate a solid correlation between Ea and GHG dynamics (concentrations and emissions). This was not always clear to me in the discussion.

              Furthermore, it is not clear to the reader, for example, how sediment characteristics (what is meant by this?) and biological communities (which are referenced here) affect the GHG cycle. What information is contained in the references and how does it relate to GHG dynamics? Since, in my opinion, this is central to the strategy of the study, it needs to be introduced here and addressed in more detail in the discussion.

L77ff     The postulated correlation between chl/nutrients (TP) and open/closed bay is not clearly evident in the table.

Fig. 1     This is the first time I have seen a TomTom map in a publication. If you want to keep it, it needs to be revised: the cities are not readable and some of them can be deleted. The same applies to the roads. The land color is very weak and could be better distinguished from the water.

L86        At a depth of 30 cm, it is really possible to avoid air bubbles being sucked into the WEGAS system. What were the ship's speed and wind conditions (waves)?

L90        How was the system calibrated?

L95        How were the ppm determined? Why are CO₂ concentrations not converted?

L98        How were the partial pressures determined? How was the total pressure in the Equi headspace determined?

L104      Eq 3. In which calculation is the Bunsen coefficient used? Please complete the equations.

L106      How and where was the temperature measured?

L106ff   “For N2O, K0 = β. For CH4 (ideal gas behavior), K0 = β(R×273.15 K)”, Please write complete and understandable sentences.

L109      Please write “air-sea fluxes of GHG ...”

L111      “K₀ is solubilty”, the connection to Eq. 3 is not clear here. Complete the equations.

L12        The usability of this k-model must be explained. Why is it considered suitable? And does it allow for comparison with published fluxes (Table 5)?

L116      Sc_{this study}, “this study” can be deleted as it is not mentioned in Eq 5.

L117ff   Is it really appropriate to use an average value here? Wind events prior to the measurements could have caused increased GHG emissions from the water. This may make it difficult to compare the bays and time periods. The wind data for the measurement days should be shown here. How does the value of 2 m s-1 fit in with the measured data? Please discuss.

L125ff   This section on methodology lacks any references.

L139      Were all basins inspected by divers a few weeks before September? Please write more clearly in the text. How were these estimates of vegetation cover (in %) made? There is no reference to the method used here.

L142      “The interannual variation in total vegetation cover is rather small in these bays”, How do you know that? Please provide a reference.

L144ff   The word “plot” seems a little unusual to me. Maybe replace it.

L146ff   Why were these two vegetation indicators chosen? Please explain.

L150ff   This section on methodology lacks any references.

L154      Is the sediment dry density of 2.65 g cm-3 (Burdige et al.) applicable to all sediment types in the different bays (inner and outer part of the bay)?

L157ff   Is it really the case that the top 1 cm of sediment is responsible for GHG dynamics in surface water? Methane production usually occurs in deeper sediments. And what about possible gas releases from deeper sediment strata? How does the top 1 cm of sediment relate to this?

Results and Discussion

L161      This entire paragraph refers to important findings that are only indicated here. Instead, reference is made to the statistics in the appendix. Since this is a central part of the manuscript, the results must be included in the main text and explained in more detail (including in a graphical representation, not just in tables). What does it mean that statistical tests “confirmed significant differences in GHG concentrations between bays”? Wouldn't one expect that? What are the differences? How pronounced are the differences and do they allow for classification in the Ea category? Can environmental drivers that influence GHG concentrations be derived from this?

Fig. 2     This graphic needs to be revised: (a), (b)... need to be aligned with each other, made larger and bold; lines in the illustrations are too narrow and unclear and could possibly be removed entirely. The scaling of the y-axis also varies, which makes comparison difficult. I suggest sorting the bays into “closed” and “open” so that the reader does not always have to refer to Table 1. Why are CO₂ concentrations given in ppm (in headspace gas) and not converted to CO₂ dissolved in water?

L167      I would remove the word “pattern” from the headings.

L168      What are the saturation values referred to here? Perhaps you could mark them in Fig. 2.

              Why is reference made to Table 2 and A1 and not to the central Figure 2?

L170ff   “These bays showed significantly higher CO2 concentrations inside compared to outside areas” Given the overall inconsistency in Fig. 2 between all bays, I'm not sure if this really needs to be pointed out. The discussion here should focus more on why such an inconsistent pattern occurs (in contrast to other GHGs).

L175      How can the high CO2 concentrations and high degree of eutrophication at BV be explained? This contradicts the statement made in the text.

L176      “counter-intuitive that the least eutrophied … bays”, BV indicates a comparatively high degree of eutrophication in TP and Chl. So how does the statement in the text match up with this?

L177      “this might be due to high allochthonous, terrestrial input and autochthonous input from decaying plant matter”, What is this assumption about these bays based on (references?), and why should allochthonous input into these bays be different from the other bays studied?

L188      How does the author conclude that this coastal CO2 source is overlooked? It is understudied. What do other studies say about this topic (references)?

L193      What are the saturation values referred to here? Perhaps you could mark them in Fig. 2.

              Why is reference made to Table 2 and A1 and not to the central Figure 2?

L196      What is the context of the reference to Conrad (2009)? Is this about high methanogenic activity due to the conversion of organic material?

L198      What is meant by “degraded system”?

L201ff   Reference missing to studies that have already shown this.

L202ff   Figure 2 does not show that closed and open bays can be clearly grouped according to their CH4 concentration. Rather, the picture is inconsistent. I thought that the author's intention was to be able to clearly assign GHG concentrations to open and closed bays, but this is difficult to see. Please reconsider your statement.

L205ff   How does this statement fit with your own data: In Table 4, OCsed (positive CH4 conc.) correlates with vegetation cover (negative CH4 conc.). Please discuss using your own data! Reference to the statement at the end of the sentence is missing.

L211      Please replace the word “boost.”

L211ff   Warming of the water column leads to stratification and high CH₄ concentrations below the thermocline and low concentrations above the thermocline (exchange with the atmosphere).

L213ff   An illustration showing the correlation between conc. CH4 and salinity is missing here. Where does the fresh water come from? Do the bays differ in terms of their river inflows?

L216ff   There is no apparent connection between the sentences. Why does this sentence have to be here? It has no connection to the explanations in the previous sentences.

L217      Reference is missing.

L218ff   I would move this paragraph up to where CH4 concentrations were discussed already (L193ff).

L218      How do these values compare with those for the open Baltic Sea?

L220      Where were the studies conducted that are cited (Humborg and the other studies)? All studies in Tvärminne?

L226      What are the saturation values referred to here? Perhaps you could mark them in Fig. 2.

              Why no reference to Table 2 and A1 like above and not to the central Figure 2?

L228      Only one enclosed bay (HV) is shown in Figure 2e. Can we really then speak of a “clear pattern”?

L233      A reference to temperature-dependent enzyme activity is missing.

L234ff   Where is the correlation between NO2+N03 and N2O shown? Why is it different in the bays? Formulate hypotheses and discuss them.

L236ff   What is meant by “flow rates”? Please clarify!

              Why does this have an impact on denitrification? Please explain and provide references.

L238      Again, “flow rates.” Does the author mean “currents”? Please clarify!

              “creating different biogeochemical conditions”, creating different biogeochemical conditions, to what end? Please discuss why this has an impact and what kind of impact.

L238ff   “N2O production…”, That's a little out of context again. I would suggest incorporating that into the introduction, where the processes have already been described.

L242ff   Please establish a clearer link to the studies listed and discuss the differences/similarities (e.g. correlation with O2 concentrations). Can the study by Brunberg (2025) already be cited, as it is still under review? Why does the summer bloom not lead to a comparable N2O signal?

Tab. 2-4 The results need to be presented differently. I recommend a graphical representation of the key results. Otherwise, the reader will be overwhelmed by the overloaded tables.

L249      Correlation between N2O and CH4. The entire chapter is written in a very unstructured manner. I recommend reconsidering this and introducing a logical, sequential structure.

L253      The word “flow” is unclear.

L254      “high-energy environments”, this needs to be expressed more clearly. What does the author mean?

              This similarity to earlier studies should be discussed in more detail.

L256      The last sentence is a repetition of the first sentence in the paragraph. It can therefore be deleted (again, the word “flow”?).

L260      “indicative of the complex redox dynamics of these systems”, what does the author mean by this? Please discuss the assumed processes/relationships that could be important in this context.

L261ff   “Importantly,…” I don't understand the meaning of this sentence. Where can I find data on this (e.g., on the sediment properties in the bays studied)?

L263      I would recommend merging the paragraph with the paragraph in L252.

L264      The term “oxygen minimum zones” may be unsuitable for shallow water studies. Please reduce the number of references.

L267      “91%”, please explain in more detail what these values are.

L268ff   O2 was measured after all. Where is there a figure showing whether O2 and CH4 correlate with each other?

L270ff   Where are the local N2O peaks shown (figure)? Again in the appendix?

              Wind induced mixing can also stir up water that is low in O2 and rich in N2O from the water body near the seabed?

L278      What is the significance of N2 fixation in relation to your own data? Does this need to be pointed out here?

Conclusion

L315      “This study provides a comprehensive assessment of GHG emissions…”, “comprehensive” is perhaps the wrong word, as this is only a very limited study in selected basins.

L321      What is meant by “generally dominated fluxes”? What kind of flux? Please rephrase.

L322      “N2O showed opposite seasonal trends with higher concentrations outside the bays”…and lower concentrations in fall. Please complete the statement.

L323      This statement is still very weak and needs to be further extended in the discussion (in 3.1.4). How can this be explained? What does it indicate? Why is it only noticeable in some basins? Ideas should be developed here (or better in the discussion section).

L326ff   “insights into the potential environmental controls on coastal GHG dynamics”, This is where the manuscript's weakness lies. The “controlling drivers” need to be better identified and presented. A clear classification is not apparent in the text.

L328      “high spatial … variability”, does this refer to spatial variability in individual basins (e.g., inside/outside bays)? To this end, figures from the appendix (e.g. Fig. A1) should be transferred to the main text and discussed in greater detail (perhaps only for selected bays).

L332ff   “overlooked but important component”, “overlooked” is probably not the right word. We know that they are important. If so, they are undersampled (spatial heterogeneity) and the processes that control greenhouse gas dynamics are poorly understood.

L336ff   How can “EC flux measurements and monitoring of seawater” help to better identify the biological drivers (methanogenic and methanotrophic; both in seawater?)? Please rephrase.

L338ff   Please include the problem of not capturing gas bubble fluxes in the discussion and discuss it based on existing literature (e.g., Bisander et al. 2025).