the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 30 Jan 2025
Aquatic metabolism influences temporal variations of water carbon and atmospheric carbon dioxide fluxes in a temperate salt marsh
Abstract. Salt marshes are blue carbon (C) ecosystems characterized by intense atmospheric CO2 uptake and C sequestration but also organic and inorganic C exports through the tide. However, uncertainties on main biotic factors controlling vertical and horizontal C fluxes imply studying simultaneously terrestrial and aquatic metabolisms at small timescales (diurnal and tidal) and distinguish their contributions to net ecosystem CO2 exchanges (NEE). Within a temperature salt marsh, four sampling 24-h cycles were performed to measure water biogeochemical parameters (carbon and nutrients) and planktonic metabolism simultaneously to NEE successively at high tide (imported coastal waters influenced by the continental shelf) and low tide (exported channel waters influenced by the marsh). At high tide, water CO2 oversaturation due to aquatic heterotrophy was able to significantly reduce marsh atmospheric CO2 uptake at the ecosystem scale (NEE) during the highest immersion levels. At low tide, water pCO2 were also mainly controlled by marsh biological activity inducing large water CO2 oversaturation in winter due to heterotrophy and large water CO2 undersaturation in spring and summer due to autotrophy. In winter, the highest increases of dissolved inorganic carbon (DIC; from 2354 to 3963 µmol kg-1), total alkanity (TA; from 2508 to 4016 µmol kg-1) and dissolved inorganic nitrogen (DIN; from 27.7 to 68.4 µM) were measured at low tide night probably due to intense anaerobic respiration processes in channel waters and/or sediments resulting in the highest water pCO2 (up to 1461 ppmv). On the contrary, in spring and summer, large water pCO2 decreases and dissolved organic carbon (DOC) increases from high to low tide could be related to intense autochthonous and allochthonous aquatic primary production. Over the 24-h cycles, planktonic metabolism strongly influenced water pCO2 variations, especially at low tide, though planktonic communities did not play a major role in the atmospheric C balances at the ecosystem scale (NEE), accounting for only 10 % in spring.
- Preprint
(1566 KB) - Metadata XML
-
Supplement
(588 KB) - BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2025-335', Anonymous Referee #1, 14 Mar 2025
Dear Authors,
This study entitled “Aquatic metabolism influences temporal variations of water carbon and atmospheric carbon dioxide fluxes in a temperate salt marsh” has several distinct strengths. Firstly, the author systematically observed changes across four seasons—spring, summer, autumn, and winter—and further examined diurnal variations within each season over 24-hour cycles, successfully capturing seasonal and daily fluctuations in carbon dioxide fluxes. Additionally, the author employed multiple measurement methods, including changes in atmospheric CO₂, water partial pressure of CO₂, nutrients, planktons, and biological parameters. However, further clarification of the study's scope, limitations, and implications will substantially enhance the research's logical coherence and completeness. My major comments are as follows.
1. The author emphasizes the study of "ecosystem carbon dioxide exchange" in the introduction, but the main discussion of the paper actually focuses on the water interface. The authors also used "air-water CO₂ exchange" later for the measurement in the water. Further, NEE and air-water gas exchange actually measured at different locations. The authors can use a conceptual model to organize the role of each in this study area and also indicate the physical or theoretical boundaries and limitations of this study.
Upon closer inspection of the methods and results, the study shows unclear definitions regarding its spatial scope and scale. The author claims to have conducted "vertical and horizontal surveys," yet in practice, the vertical survey is essentially limited to the air-water interface. The so-called horizontal survey is restricted to observations from a single point, where the author assumes tidal movements bringing upstream and downstream waters into the sampling area over a 24-hour period suffice as horizontal analysis. This approach, however, is highly limited since single-point sampling cannot represent the actual spatial variations throughout the salt marsh region. What is the role of this study site on this whole area, including the evaporation ponds all over the island?
Consequently, the observed data might disproportionately reflect sedimentary and anoxic environmental influences from the upper or lower stream rather than the actual diurnal variations caused by photosynthesis and planktonic activity. Can this single selection inadvertently led the author to interpret tidal-driven signals as representative of the entire salt marsh ecosystem?
To address these issues, the author should clearly define the limitations of the study, explicitly describing the representativeness of the sampling points within the salt marsh area. It is crucial to specify under which spatial conditions the observed results are applicable, distinguishing clearly between areas with longer or shorter water flow paths, and between flowing or stagnant water bodies. Furthermore, the author should further explore how air-water CO₂ fluxes are influenced by temperature and wind speed variations under different seasonal and diurnal conditions, and clearly state which factor has the more significant impact.
2. The role of mixing. Moreover, since the study site is located at the river-sea interface, the dynamics of water mixing should be investigated in greater detail. The author should discuss how water mixing processes affect the study results, thereby enhancing the regional significance of the research.
3. A few sentences can be modified to improve the reading.
Line, 675-676. The authors may change the sequence of presentations for the season. For example, “inducing water CO2 undersaturation in spring/summer and water CO2 oversaturation in fall/winter.” This may be applied to the entire article.
Line, 682-684. This sentence is hard to read and can be spectacular. The Reviewer suggests splitting this sentence into two sentences and clarifying each sentence. Similar sentences can be found in the abstract, making this sentence difficult to interpret.
4. Uncertainties induced by k is unclear, Line 257. The authors should justify the reason to use this k in one or two sentences.
5. Implication: Most importantly, it is suggested that the author explicitly articulate the global or regional implications of this research clearly in the final sentence of the abstract to underscore its significance. Since this study employs both atmospheric and aquatic CO₂ measurement methods, the author should consider comparing and discussing the differences and relationships between these two measurement methods, potentially providing predictions or assessments regarding regional variations. Establishing a clear conceptual model based on such comprehensive and systematic observations would greatly enhance the research's academic value and influence.
Minor comments:
Regarding presentation, it is recommended that the author rectify the inconsistent font sizes in the figures. For example, the term "pCO₂" is excessively large while other text is too small, causing readability issues. All text should be consistently sized and easily readable. Additionally, the labeling of "pCO₂" should remain uniform throughout the paper.
Citation: https://doi.org/10.5194/egusphere-2025-335-RC1 -
RC2: 'Comment on egusphere-2025-335', Anonymous Referee #2, 17 Mar 2025
Title: Aquatic metabolism influences temporal variations of water carbon and atmospheric carbon dioxide fluxes in a temperate salt marsh by Mayen et al.
General Comments:
This manuscript presents valuable insights into the metabolic dynamics of a temperate salt marsh ecosystem, with a focus on temporal variations in DIC and CO₂ fluxes. The study is thorough in its approach and provides meaningful data across seasonal and tidal cycles. However, addressing the points below—especially those related to external influences, methodological clarity, and data interpretation—will improve the manuscript’s clarity and impact.
Contextualization with broader literature:
While the study provides localized data on carbon dynamics, its relevance to broader regional or global trends is not fully explored. Including references to similar studies or integrating global estimates would help contextualize the findings and enhance the manuscript’s scientific significance.
Discussion of study limitations:
The manuscript would benefit from addressing potential limitations, particularly the absence of direct measurements of anaerobic respiration. Since both aerobic and anaerobic respiration pathways are key drivers of DIC and TA generation, a discussion on the lack of direct data and its implications would improve the transparency and robustness of the study.
Influence of external inputs and confounding factors:
The potential influence of riverine and anthropogenic inputs—especially given the proximity to upstream dykes and salt ponds—is not thoroughly discussed. River inputs can introduce organic and inorganic materials, impacting gas concentrations and leading to supersaturation conditions for O₂ and pCO₂. As noted in Line 515, transient tidal phases cause considerable pCO₂ variability due to lateral exchanges. More explicit discussion on these confounding factors is necessary to support the conclusion that observed variations are predominantly driven by marsh primary production.
Role of emergent vegetation:
Section 4.2 focuses heavily on planktonic primary production, yet emergent vegetation such as Spartina maritima likely plays a significant role in carbon cycling. Data in Tables 2 and 3 suggest that marsh plant metabolism (NEEmarsh) has a greater influence than planktonic NEP. The manuscript should provide a more detailed discussion on how emergent plants contribute to DOC production and GPP.
Clarity through visual aids:
A conceptual figure illustrating the pathways of TA, DIC, and DOC generation—and their respective effects on pCO₂—would help clarify the interactions and enhance the reader’s understanding of the discussed processes.
Specific Comments:
L95: Provide sediment type information at the study sites, which is relevant for interpreting diagenetic processes. This can be added to the Methods or Results section.
L160: Standardize the y-axis scaling for CO₂ in Fig. 2 to ensure comparability across plots.
L178–179: Clarify whether the listed plant species percentages are based on biomass, coverage, or another metric. Also, indicate whether vegetation composition varies seasonally, and how this might impact CO₂ fluxes.
L194–195: The reported uncertainties are appreciated. Specify whether they were determined through replicate analyses, CRM comparisons, or another quality control method.
L245: It is noted that water samples were collected for DIC analysis, yet DIC concentrations were calculated rather than directly measured. Clarify this apparent inconsistency.
L270: The formula presented appears incorrect. (TA1 - TA2) should be multiplied by a factor of 0.5, and DIC and TA should be referred to as NDIC and NTA.
L530: The manuscript discusses seasonal shifts in metabolic status, but the role of light (PAR) across seasons and tidal phases is underexplored. If PAR data were collected, integrating it into the analysis would enhance the interpretation.
L540: The use of POC stable isotope ratios is mentioned; briefly explain the methodology and analysis in the Methods section.
L675: The statement “Over the seasonal 24-h cycles, water pCO₂ was mainly controlled by biological activity...” is not supported by the data. Fig. 2 indicates that pCO₂ variation corresponds closely with Hw, suggesting tidal forcing as the main driver. This should be addressed.
Figures and Tables:
Fig. 2:
In winter, high salinity occurs during high Hw, whereas in spring and fall, high salinity coincides with low Hw. Additionally, no clear relationship between salinity and Hw is observed in fall. These seasonal differences warrant further discussion. Moreover, pCO₂ does not exhibit significant diurnal variation during low Hw periods, which diverges from typical diurnal patterns. This anomaly requires explanation.
Table 1:
The highest Chl a concentration is reported in fall, which contradicts the data shown in Fig. 4. Clarify this inconsistency.
Fig. 4:
Chl a concentrations are high in fall, but both microphytoplankton and pico-nanophytoplankton abundances are low. This discrepancy should be addressed in the discussion.
Fig. 5a and Table 2:
NEPpK > 0 is shown during Ht/Night in spring; however, NEPpK cannot logically be positive during nighttime. An explanation for this abnormal result is necessary.
Fig. 6:
TA and DIC show a negative correlation with salinity in winter and fall, but a positive correlation in spring and summer. The reason for this seasonal shift needs to be discussed.
Citation: https://doi.org/10.5194/egusphere-2025-335-RC2 -
RC3: 'Comment on egusphere-2025-335', Anonymous Referee #3, 17 Mar 2025
The main goal of this paper is to show that tidal floodwater when over the marsh alters CO2 gas flux between the tidal marsh and the atmosphere, that the biogeochemistry of the tidal water is altered when it floods and drains from the marsh, and that the CO2 flux magnitude and direction from flood water when it totally inundates all marsh vegetation is controlled by CO2 levels somewhat controlled by pelagic metabolism. The approach appears to be designed primarily just to prove these points – not to quantify the seasonal/annual pattern of any of the processes that can contribute to these patterns. I say this because water chemistry was only sampled on 4 days during an entire year (seasonally) and with the exception of pCO2, which was measured by continuous probe, other parameters (nutrients OC, IC, etc) were sampled only twice on either side of high tide and low tide. The interval between samplings at high and low tide were not mentioned, best I can tell.
The sampling conducted here was done in a tidal creek that connects a marsh to open bay water and the ocean. The creek and sampling point were within the footprint of an ongoing eddy covariance tower study. However, the footprint of the tidal creek and its flooding watershed were never indicated. Indeed, when I look at a google earth image of the study site, it seems to be an extremely human altered marsh/estuary, with ponds, ditches, spoil banks, etc all over the place. The flooding pattern seems impossible to describe by looking at an aerial image.
As in most estuaries there seems to be strong spatial gradients in water chemistry (and metabolism) with two end-members – way into the marsh and the ocean/bay. The authors took 2 samples at low tide, after water had drained from the marsh surface and apparently sat around still in a tidal creek until the next tide came in. why the water level never drained completely such that water level went to zero was never mentioned. The water at low tide was sampled presumably about ½ way before the current went slack and then picked up again. Water was sampled again at high tide – presumably, the water was in the bay only minutes before and therefore showing little to no immediate effect of being on the marsh (of course the estuarine water at high tide presumably had been influenced by previous flood tides – but we don’t know what the water residence time was and what its time course trajectory was. We also don’t know how deep the water was when it flooded the marsh during the different times samples were collected. We don’t know if the vegetation was completely flooded or not. We don’t know if the tidal creek was influenced by the many ponds on the marsh which apparently had dense macroalgae. We do know that there was extensive salt accumulation in the summer, presumably reflecting evaporation in the many ponds, which may or may not have flooded and drained on a regular basis.
We learned – that the flux of CO2 from the tidal marsh to/from the marsh surface and its plants was shut off at high tide. We also know this from just about every flux tower study that’s been conducted that I’m aware. So nothing new was learned here.
We learned that the remaining CO2 flux from the marsh surface when it was completely covered by water reflected not processes occurring in/on the salt marsh, but metabolism of the water itself. This means that the pCO2 levels were entirely due to the metabolism – planktonic and benthic, which were not completely separated. Light-dark bottles were used to simulate the pelagic metab and everything else plus plankton were measured by the time course change near high tide or low tide. However, the water at high tide wasn’t over the marsh – it was in the tidal creek. So it doesn’t reflect the marsh surface sediment fluxes, as they authors didn’t say they sampled water out in the middle of the marsh after it had traversed the marsh surface flooding away from the tidal creek. While light-dark bottles can give some measure of pelagic metabolism, were they suspended at multiple light levels – I don’t think so. Was shading from marsh plants imitated – I don’t think so. There was some metabolism as the authors documented it. The level of pCO2 in the floodwater was related to the direction of flux seen by the flux tower. When floodwater at high tide was supersaturated – there was an escape of CO2 to the atmosphere and vice versa.
We learned therefore that flux direction and magnitude from a marsh can change when the marsh is flooded. But we don’t know whether the magnitude calculated bears any semblance to reality and how it really compared to what the flux tower saw. Actually it could have been interesting to see what k value should be used to calculate flux from the water surface when it is over the marsh, as the flux tower doesn’t rely on a k. Presumably the k value at a particular wind speed differs when the marsh vegetation sticks above the surface or alters water turbulence. This is a big unknown, but unfortunately the authors didn’t measure this. I don’t think we really learned by comparing light-dark incubations with whole water DIC changes the true contribution of pelagic vs all other processes that control DO or DIC concentrations in a water column in contact with sediments, plants, the air, etc.
We also learned that aquatic metabolism isn’t always greatest in the hot summer, as this study showed the highest and greatest range in DOC during the winter – the coldest month. But of course this result wasn’t strictly a function of temperature.
An extensive set of observations was presented here. The authors justified many flux trajectories by mentioning one of many processes previously studied in marshes. For example, they showed that through a particular DIC-DIN flux ratio and NH4:NO3 ratios that there may have been DNRA. They also showed via the slope of the DIC/TA change could have been explained by sulfate reduction. But they showed it being greatest in winter, when there is often less sulfate reduction and more pyrite reoxidation – changing the O2 to DIC flux ratio of metabolism.
While there is a lot of data presented here – I see little that’s new. It’s from an extensively altered salt marsh so extrapolation of results to anywhere else is dubious – except for saying some of the same marsh processes occur there as everywhere else. I don’t see that we are any closer to understanding the magnitude and direction of DIC flux to/from a regularly flooded marsh. Nor are we any closer to quantifying the CO2 sequestration rate of this salt marsh system.
Citation: https://doi.org/10.5194/egusphere-2025-335-RC3
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 166 | 55 | 9 | 230 | 22 | 7 | 13 |
- HTML: 166
- PDF: 55
- XML: 9
- Total: 230
- Supplement: 22
- BibTeX: 7
- EndNote: 13
Viewed (geographical distribution)
| Country | # | Views | % |
|---|---|---|---|
| France | 1 | 71 | 33 |
| United States of America | 2 | 58 | 26 |
| China | 3 | 15 | 6 |
| Taiwan | 4 | 11 | 5 |
| Brazil | 5 | 7 | 3 |
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
- 71