The Alkalinity Blue Carbon Pump: Tidal pumping drives long-term CO₂ sequestration as alkalinity in salt marshes

Kanfer, Paul; Charbonnier, Céline; Schmidt, Sabine; Deflandre, Bruno; Mouret, Aurélia; Bénéteau, Eric; Danilo, Martin; Poirier, Dominique; Desnues, Jimmy; Lebreton, Gérard; Anschutz, Pierre

doi:10.5194/egusphere-2026-5782

Preprints

https://doi.org/10.5194/egusphere-2026-5782

Preprints

01 Oct 2026

| 01 Oct 2026

Status: this preprint is open for discussion and under review for Biogeosciences (BG).

The Alkalinity Blue Carbon Pump: Tidal pumping drives long-term CO₂ sequestration as alkalinity in salt marshes

Paul Kanfer, Céline Charbonnier, Sabine Schmidt, Bruno Deflandre, Aurélia Mouret, Eric Bénéteau, Martin Danilo, Dominique Poirier, Jimmy Desnues, Gérard Lebreton, and Pierre Anschutz

Abstract. Tidal salt marshes are increasingly recognised as significant blue carbon ecosystems. However, their total carbon sequestration capacity remains underestimated because dissolved carbon outwelling and alkalinity export are rarely considered alongside sediment burial. Here, we combine tidal creek time series measurements, open incubations, and sediment core analyses to quantify all the main carbon flux pathways in a temperate macrotidal salt marsh of the Arcachon Bay (SW France). A binary salinity mixing model coupled to direct discharge measurements allowed us to isolate the net tidal pumping fluxes of total alkalinity (TA), dissolved inorganic carbon (DIC), dissolved organic carbon (DOC) and methane (CH₄) from conservative mixing. DIC outwelling (5.3 ± 1.7 mmol m^-2d^-1) was the largest form of dissolved carbon export but it does not fully represent net sequestration as part of it consists of CO₂ that returns to the atmosphere via air-water exchange. Net TA lateral fluxes averaged 3.9 ± 1.4 mmol m^-2 d^-1, driven by sulphate reduction coupled to pyrite formation and aerobic respiration in marsh sediments, as supported by TA:DIC ratios and δ¹³C-DIC. DOC outwelling (3.4 ± 3.3 mmol m^-2 d^-1) was on same magnitude but its sequestration potential depends on lability. High tide benthic incubations revealed significant TA and DIC fluxes from sediments to the water column, higher in vegetated than in bare mud areas, driven by benthic processes and root-network macroporosity. Organic carbon burial rates estimated from ²¹⁰Pb_xs profiles were around 80 mmol m^-2d^-1 in vegetated zones. However, these rates likely represent an overestimated gross burial rather than net sequestration given the dynamic and erosive nature of this macrotidal system. When integrating all measured carbon flux, net TA outwelling emerges as the most chemically conservative and durable sequestration mechanism, as it is resistant to remobilisation and can remains stable over millennial timescales once exported to the coastal ocean. We propose the term Alkalinity Blue Carbon (ABC) to describe this pathway, and call for its systematic inclusion alongside sedimentary C burial in future blue carbon inventories and climate mitigation frameworks.

Received: 22 Sep 2026 – Discussion started: 01 Oct 2026

Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.

Download & links

Paul Kanfer, Céline Charbonnier, Sabine Schmidt, Bruno Deflandre, Aurélia Mouret, Eric Bénéteau, Martin Danilo, Dominique Poirier, Jimmy Desnues, Gérard Lebreton, and Pierre Anschutz

Status: open (until 27 Nov 2026)

Post a comment Subscribe to comment alert

RC1: 'Comment on egusphere-2026-5782', Anonymous Referee #1, 20 Oct 2026 reply

See attached

Reply

Citation: https://doi.org/10.5194/egusphere-2026-5782-RC1
RC2: 'Comment on egusphere-2026-5782', Anonymous Referee #2, 23 Oct 2026 reply

Overview
The paper “The Alkalinity Blue Carbon Pump: Tidal pumping drives long-term CO₂ sequestration as alkalinity in salt marshes” measured carbon fluxes in a small saltmarsh in France. It includes seasonal time series during low tide, mixing models, open incubations and sediment cores. The dataset is very comprehensive, the manuscript well-structured and figures informative. However, there are some mythological issues that need to be resolved before publication.
General comments
Reading the abstract, I assumed you measured pyrite and aerobic respiration. Only say “driven by” if you measured the driver, if not add “potentially” or remove. The abstract (also the results and discussion) should focus on the strengths of the paper: seasonal changes and biogeochemical drivers, rather than the weak points (ecosystem scale fluxes).
The introduction is quite broad. It could be more tailored to the study, clearly defining specific knowledge gaps. The last paraph should include a summary how research questions were addressed.
Calculating daily fluxes based on only ebb tide, using a freshwater endmember from a different water body and ocean endmember from a mid-tide saltmarsh water sample seems problematic. I suggest presenting results for high tide and freshwater endmembers and a mixing model with exclusively ebbtide data (add a figure showing fluxes per month for both). Then emphasize limitations and what each approach represents. Giving total fluxes per hour/ low tide seems more appropriate than daily fluxes.
Using open and static incubations to quantify process that are driven by tidal pumping in deep anoxic layers likely underestimates TA and DIC fluxes.
Expand sections in results and discussion about seasonal trends and biogeochemical drivers of TA and DIC (title suggest these are your focus). E.g. add some regressions of concentrations or ebbtide delta values. Largely reduce and summarise sections about outwelling and fluxes (remove Fig 8).
Comments by line
L16 Ta fluxes driven by aerobic respiration?
L19 How much higher?
L161 And DIC?
L222-229: On the maps it lokes like the creek and the North Canal are not connected. Are there some small canals connecting them or how can you justify applying a mixing model with endmembers from a different water body?
L283 This section is not central to your overall story and could be presented as a supplementary table (and text) instead. These values are only relevant for the mixing models.
Fig 3. Delta TA, DIC and DOC continue to increase until 240. This suggests that the maximum delta concertations have not yet been reached potentially underestimating tidal fluxes.
L353-361 The mudflats are likely also TA sources. Using a mid-tide sample as ocean endmember seems wrong to me and explains the extremely low carbon fluxes (e.g. 5 DIC mmol/m2/d) versus other saltmarsh sites with typically > 100 mmol/m2/d. I would present results for high tide and freshwater endmembers and a mixing model with exclusively ebbtide data. Then emphasize limitations and what each approach represents.
Fig 4. Also show nutrients and DO.
Fig 6. Use delta only from ebb tide data.
L772 Subscript.

Reply

Citation: https://doi.org/10.5194/egusphere-2026-5782-RC2
RC3: 'Comment on egusphere-2026-5782', Anonymous Referee #3, 06 Nov 2026 reply

Summary
The study discusses an important topic in blue carbon science: carbon and alkalinity outwelling from coastal wetlands. The study presents estimates for a few important carbon flux pathways in a salt marsh, including lateral exports of carbon and alkalinity, vertical air-water fluxes of greenhouse gases, vertical exchange of porewater over the inundated mudflat and salt marsh, and soil organic carbon burial across the platform. It presents data from sampling that occurred across a full annual cycle in a temperate climate, which is important for resolving seasonality given the scarcity of wintertime or cold season lateral export data in the literature. However, I have serious methodological concerns with the outwelling estimates. I also find the title and general framing of the study as evidence for long-term carbon sequestration through alkalinity exports to be misleading.
Major Comments
I have concerns over the reliability of the new method presented for quantifying carbon and alkalinity exports due to the potential for large, unreported or unquantifiable uncertainties. The method is similar to previous time series approaches to quantify fluxes over individual tidal cycles which employ discrete sampling of chemical constituents and tidal water flow. However, this method employs sampling over only part of an ebb tide to quantify fluxes, assuming that the chemistry of incoming tidal water will be governed only by conservative estuarine mixing. Some previous studies which have used a similar mixing model approach to quantify “excess” solutes in tidal waters have sampled across full tidal cycles and demonstrated that flood tide waters approximately follow estuarine conservative mixing. Because this is not done here, we have no way of knowing whether the flood tide water is represented by the chosen endmembers, which could be introducing large uncertainties. Based on the data presented in Fig. B1, there is not clear evidence of conservative mixing throughout the estuary, as the three potential marine endmembers given (offshore, inlet, and first time series sample) do not all fall on the same mixing line with the chosen fresh endmember. The endmember that is actually used for the ΔTA, DIC, etc. calculations is ebb tidal water near high tide, and typically plots above the other mixing lines shown, sometimes considerably. Therefore, it appears that this water is likely already marsh-influenced (i.e. contains porewater derived carbon and alkalinity). This may be contributing to the very low outwelling values reported here compared to many other studies, as this choice of endmember may be reporting deviations from conservative mixing that are smaller than an actual flood tide endmember that is not marsh-influenced would produce.
In general, a marsh tidal creek is likely more accurately represented by a three-endmember mixing model that includes an upstream fresh endmember, a downstream saline endmember, and a marsh porewater endmember, as has been done previously with radiochemical tracer approaches. It is likely that this can be simplified to the two-endmember mixing model that is presented here in many cases since at high stages, porewater will make up a small fraction of tidal creek water and therefore influence its salinity minimally. However, during a few tidal cycles, salinity actually increases from high to low tide (Fig. 2, 1B), which is counter to the expected two-endmember mixing, likely due to contributions of saline porewater at low tide. In these cases, the two-endmember mixing model may fall apart completely.
Other uncertainties may arise from the fact that the complete ebb tidal cycle is not sampled, as sampling only begins after the stage has fallen below a certain level. Because of this, there may be a large but unquantifiable volume of outgoing water between high tide and the start of sampling that is not represented. This may also be contributing to the very low flux values, since as this would be a very high discharge period, export during this time may be considerable. I do not see high tide stages reported for the chosen tidal cycles, but I would assume that stage and therefore discharge should be highest during spring tides, moderate during “mid” tides, and lowest during neap tides. However, this is not always the case, as “mid” tides at times have greater discharge than spring tides (Table 2), which may mean that the sampling approach causes a somewhat arbitrary discharge volume to be reported, which will impact reported fluxes with unquantifiable uncertainties. In general, because there is no sampling during the flood tide, there is no guarantee that the reported discharge values represent “balanced” water budgets (i.e. equal incoming and outgoing flow), which could again lead to major uncertainties in the reported fluxes. There is also no way to assess whether the sampled tides are representative of their given month, which may not be the case given that fluxes over individual tidal cycles have been previously demonstrated to be highly variable. This can contribute to uncertainty when upscaling these single tidal cycle fluxes to an annual rate, but the magnitude of this uncertainty is unknown.
Another potentially large source of uncertainty in outwelling and CO₂ efflux calculations is the fact that DIC and CO₂ concentrations were not directly measured but instead calculated from TA and pH. This can be caused by analytical uncertainties in input variable measurements (particularly with probe pH values) but may also be the result of the presence of non-carbonate or organic alkalinity species in marsh tidal waters. In general, the manuscript lacks robust analyses of these potential uncertainties. Some sources of uncertainty can likely be estimated, but others may not be able to be adequately quantified with the available data.
Another major issue with the manuscript is that it equates alkalinity outwelling to long-term carbon sequestration. However, the actual data presented does not address the question of whether alkalinity outwelling can serve as a durable carbon sequestration (i.e. whether alkalinity exports remain conservative across the estuarine and shelf systems to reach the ocean). I agree with the authors that this is likely the case based on some of the previous work cited but feel that it remains an open question in carbon cycle science the extent to which it is true and that the alkalinity additions drive long-term carbon sequestration. Therefore, I do not feel that the study provides evidence related to this question simply by showing that alkalinity is outwelled from a marsh and so it should not make as strong claims related to carbon sequestration or present these claims as novel. Similarly, some of the conclusions focus too much on these interpretations rather than the actual results presented in the study. The study actually reports alkalinity exports that are very small compared to soil organic carbon burial. The authors propose inaccuracies in their reported SOC burial rates; however, their reported rates are in line with previous estimates for salt marshes. Conversely, their reported alkalinity outwelling is much smaller than previously reported rates, but the authors still argue the importance of alkalinity outwelling compared to SOC burial despite their results.
Comments by line
Title: Following earlier comments, I think it is too broad and not specific to the findings of the study.
L22-24: This claim is based on previous literature and not the specific findings of the study. It should therefore be removed or it should be made clear in the statement that this is the case.
L106: How was the size of the catchment area determined?
L125: Are these manufacturer-specified precision values? Actual precision in the field may be less, particularly for pH.
L160: The marsh would typically not be inundated for this long. Why was this time chosen?
L184: If no CRM was used, how was the accuracy of the in-house TA standard assessed? If the standard was prepared multiple times over the course of the study and it is not traced to CRM, changes in the standard could contribute to differences in TA values across sampling events.
L227-8: More details are needed on this calculation. The creek gas flux calculation in mol m^-2 h^-1 should not require any horizontal dimensions of the creek. This value (500 m creek distance) is probably instead used to calculate the mol d^-1 fluxes reported later, although it is unclear exactly how. Is the creek area represented by this distance and some width, and that is the area that CO₂ fluxes are calculated over? How is the 500 m distance determined? What about fluxes while the marsh is inundated that occur outside of the tidal creek?
L285: Salinity range observed at the fresh endmember indicates that it is tidal and may therefore have marsh or estuary influences. How does this affect the mixing model?
L295-308: It is somewhat confusing that the endmember discussed here is not actually the endmember used in Δ calculations to determine outwelling. This should be clarified, and those endmember characteristics should be given.
L297-8: It would be helpful to demonstrate this in a plot with regression statistics and comparison to the chosen ebb tide endmember values.
L301-2: Did DOC appear to have a conservative relationship with salinity that justifies the mixing model?
L313-316: What about discharge during mid tides?
L359-361: If the samples are from ebb tide, how did the authors assess whether they were representative of flood tide water that is not marsh influenced?
L390-391: To what extent was the difference in fluxes related to discharge vs. Δ concentration values. Did the Δ values vary seasonally? Because they tend to follow salinity over spring and neap tides, it seem like some variation is just an artifact of the chosen mixing relationships.
L463: “Marsh porewater” may make more sense to use than “groundwater” here.
L466: This section does not discuss the large body of outwelling studies that have used time series flux measurements over full tidal cycles, or the smaller but relevant group of studies that have generated high-frequency time series from extended in situ monitoring.
L478-9: What is the evidence for this statement? How do the authors know they captured variability over tidal cycles (during the same stage and season) or in porewater that was not measured?
L503-505: What is this claim of accuracy based on? There is not a thorough uncertainty analysis for this approach.
L524-5: River/canal freshwater may not be as important, but there is no discussion of the potential for fresh groundwater discharge contributing to fluxes. Based on the disconnectedness of the sampled creek from the canal endmember, to what extent could the salinity variation in the creek be the result of fresh groundwater discharge, might this groundwater have different solute concentrations than the canal endmember, and how would this impact the mixing model?
L534: This section could use better organization. The topic jumps around from outwelling to air-water fluxes and back to outwelling.
L560: The timescale of air-water CO₂ equilibration is generally thought to be relatively slow. The low pCO₂ at high tide could simply indicate water of more marine characteristics due to mixing/dilution of marsh signal and doesn’t necessarily indicate that all marsh-derived CO₂ has already been degassed.
L570: I would not consider this study to use a high-frequency approach based on sampling frequency. Clarify what is meant here.
L587-590: Several points here. (1) It would probably be better to compare reported rates specifically to other salt marshes, not just any coastal wetland (i.e. mangroves). (2) Of these three reverences, the rates listed seem to be sometimes creek-based total “outwelling” measurements from time series data, but sometimes radiochemical tracer based porewater drainage fluxes. This should probably be clarified if these are listed all together. (3) What is meant by gross fluxes? The distinction between porewater flux and overall outwelling flux is important and should be noted, but I don’t see how this relates to gross or net fluxes. How are the rates listed here that are based on time series monitoring of complete tidal cycles (i.e. quantifying fluxes in and out) not representative of net exports?
L558: CO₂ evasion is not discussed in this section but can modify the DIC:TA ratio of fluxes.
L610: The stoichiometry listed for sulfate reduction should apply to sulfate reduction coupled to pyrite formation not sulfate reduction on its own.
L612: Aerobic respiration ratio should be negative (i.e. TA consumption with DIC production).
L612-13: Agree that denitrification is not a dominant process, but can’t its contribution be more directly estimated based on the nitrate consumption that occurs in the ebb tidal water?
L621-23: Again, can’t this be more quantitatively determined?
L680-81: I don’t think it is reasonable to upscale rates recorded at one site to the global salt marsh area, particularly considering the large variability in outwelling rates reported globally.
L681-83: It is confusing that the upscaled alkalinity outwelling value presented here is considerable compared to global salt marsh carbon burial rate listed, but that the areal SOC burial rate given later from the same reference is much larger compared to the reported areal alkalinity exports. Is the difference due to wetland area used?
L685-6: It is confusing to discuss DIC and TA outwelling as separate potential sources of “long-term carbon sequestration” given that they are intrinsically connected as carbon pools. Same as in L735-7.
L718-9: I don’t think that SAR corresponds directly to apparent vertical elevation change. What about compaction/subsurface processes?
L765-6: Why is this term specific to this methodology? Couldn’t it apply to all alkalinity exports?
Technical Comments
It would be helpful to report SD along with mean where it is not included already.
More consistency in the units throughout the text would help compare the magnitude of different fluxes.
L78: Should be “characterize” and “predict”
L157: Remove “disposed” and “regulated”
L177: Should be “according to the references”
L224: Period after ref
L231: should be “estimated”
L383: should be “suggesting a weak”
L448: “implies bioturbation should be considered”?
L535: “with” seems like a typo and the sentence should be rephrased
L577: should be “depending on”
L647: remove second “from”
L733: remove second “pathway”

Reply

Citation: https://doi.org/10.5194/egusphere-2026-5782-RC3

Paul Kanfer, Céline Charbonnier, Sabine Schmidt, Bruno Deflandre, Aurélia Mouret, Eric Bénéteau, Martin Danilo, Dominique Poirier, Jimmy Desnues, Gérard Lebreton, and Pierre Anschutz

Viewed

Total article views: 674 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	BibTeX	EndNote
436	198	40	674	45	46

HTML: 436
PDF: 198
XML: 40
Total: 674
BibTeX: 45
EndNote: 46

Views and downloads (calculated since 01 Oct 2026)

Month	HTML	PDF	XML	Total
Oct 2026	387	169	32	588
Nov 2026	49	29	8	86

Cumulative views and downloads (calculated since 01 Oct 2026)

Month	HTML	PDF	XML	Total
Oct 2026	387	169	32	588
Nov 2026	49	29	8	86

Viewed (geographical distribution)

Total article views: 674 (including HTML, PDF, and XML) Thereof 674 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 10 Nov 2026

Short summary

We combine tidal creek time series measurements, open incubations, and sediment analyses to quantify the main carbon flux pathways in a macrotidal salt marsh. DIC outwelling was the largest form of dissolved carbon export; a part of it consists of CO₂ that returns to the atmosphere. Net alkalinity lateral fluxes driven by benthic sulphate reduction was the most chemically conservative carbon sequestration mechanism. We propose the term Alkalinity Blue Carbon (ABC) to describe this pathway.


Total:	0
HTML:	0
PDF:	0
XML:	0