the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 14 Jul 2026
Temporal variability of air–sea carbon dioxide exchange in the coastal water of the Baltic Sea
Abstract. Coasts are globally identified as major sinks of carbon. Still, the carbon balance in the coastal Baltic Sea remains unsolved. We analysed almost three years of continuous eddy covariance measurements of carbon dioxide fluxes in the coastal zone of the northern Baltic Sea in 2022–2024. Our results show that this coastal site acted as a small source of carbon dioxide into the atmosphere. The gap-filled carbon balances for 2023 and 2024 were 24.3 ± 0.3 g C m−2 and 19.2 ± 0.3 g C m−2, respectively. These results are comparable with earlier results by eddy covariance in the Baltic Sea. The measurements also reveal a consistent seasonal pattern of fluxes among the years, where the sea acts as a sink of carbon dioxide in April–May, as a variable source and sink in June–September, and a source in October–March. Maximum rates of monthly uptake were measured in April and the maximum emission values in November. Data gaps were filled using a random forest model which identified wind speed and sea salinity as the most important environmental co-variates of the carbon dioxide flux. Additionally, it was shown that the non-gapfilled monthly and yearly sums are offset due to lack of night-time data and thus uneven diurnal data coverage and that gap-filling is essential in calculating the yearly balances of fluxes.
- Preprint
(3440 KB) - Metadata XML
-
Supplement
(1632 KB) - BibTeX
- EndNote
Status: final response (author comments only)
- RC1: 'Comment on egusphere-2026-3945', Anonymous Referee #1, 10 Aug 2026
-
RC2: 'Comment on egusphere-2026-3945', Anonymous Referee #2, 13 Oct 2026
General comments
The authors of the paper “Temporal variability of air–sea carbon dioxide exchange in the coastal water of the Baltic Sea” present a study that uses an almost three-year continuous eddy covariance datasets to describe CO2 fluxes in a shallow coastal habitat of the Finnish Baltic Sea. The study contributes to addressing the knowledge gap regarding air-sea CO2 exchange in coastal shallow environments. The authors apply a random forest-based gap-filling approach to account for substantial data gaps, mainly from wind directions with land-influenced footprints.
Overall, the conclusions are supported by the presented data. The manuscript is well structured, clearly written, and provides important insights into diurnal and seasonal variability of coastal CO2 fluxes. I consider the manuscript suitable for publication after minor revisions. My specific comments are listed below.
Specific comments
Line 16: 2 times “only”
Line 19: I suggest rephrasing this part or leaving “blue carbon” out of the paper, since it is also not picked up in the discussion again. The statement that carbon uptake and storage in coastal ecosystems can be a nature-based solution to mitigate climate warming should be clarified. Those are natural ecosystem processes. Nature-based solutions would generally involve actions like protection, restoration, and management. I suggest briefly explaining this context or leaving it out.
Line 77: Explain or rephrase “exposed”
Line 94: This statement misses a reference. Also, if there is a significant difference in the aquatic vegetation, could you see a difference in CO2 fluxes between the two directions? This would be very interesting to disentangle, or state if there was no difference.
Line 118: Add a sentence on the calibration procedure of the gas analyser.
Line 127: Did the YSI probe measure oxygen or pH? If so, were they considered for the random forest gap filling? If not, why?
Line 158: I suggest moving this sentence to the sentence in line 161 “Increasing the measurement height…”.
Line 165: You assume the small islands do not influence the gas flux measurement because they are not vegetated. Could animals, e.g. sea birds, affect the measurements?
Line 173: “and the remaining fluxes were generally small in magnitude.” Is that a quality criterion or why is it mentioned? Is there an implication?
Line 176: “μmol m−2 s−1” make sure units are not affected by line breaks. Holding shift while typing a space prevents it.
Line 190: Reference to R and version missing.
Line 229: Adding “with” before “winter 2023-24” would help reading flow.
Line 235: You mentioned in line 101 that there are no long-term observations of ice phenology. However, in line 313 you are discussing periods of ice cover. Do you have data on the presence on sea ice? I suggest rephrasing or adding information of how you included sea ice presence in the dataset and interpretation of your results.
Figure 4: -Consider adding the xco2 graph from the supplement to the main manuscript. Highlighting the years by adding a background shade would help the reader see seasonal effects better, also in figure S2.
Figure 5: I suggest to use a uniform date annotation to the x axis. In Figure 4 you used “mm/YYYY” and in Figure 5 its “Jan YYYY”.
Line 260: You describe the result of a two-sided t test, referring to Table S1. However, the data shown in Table S1 are non-gap filled monthly means multiplied by length of the month. Please clarify which observations went into the t-test, and add the sample size, and whether a paired or independent t-test was performed.
Line 317: “Utö (27.1 g C m−2)” is this gap filled?
Line 356: “Wind-driven turbulence is widely recognised as the dominant control on air–sea gas exchange (Upstill-Goddard, 2006) and operates on the same temporal scale as flux variability.” You use the exact same sentence in Line 323. I suggest rephrasing.
Citation: https://doi.org/10.5194/egusphere-2026-3945-RC2
Data sets
Eddy covariance, meteorological and oceanographical data from Tvärminne, Finland during 2022–2024 Aki Vähä, Joachim Jansen, Kurt Spence, Joanna Norkko, Alf Norkko, and Ivan Mammarella https://doi.org/10.5281/zenodo.21135815
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 270 | 116 | 54 | 440 | 106 | 57 | 51 |
- HTML: 270
- PDF: 116
- XML: 54
- Total: 440
- Supplement: 106
- BibTeX: 57
- EndNote: 51
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
The article titled "Temporal variability of air–sea carbon dioxide exchange in the coastal water of the Baltic Sea" by Vähä et. al. is a well executed study that provides a rare long-term eddy covariance dataset from a shallow coastal Baltic Sea environment, addressing an important gap in our understanding of coastal carbon dynamics. The manuscript is clearly written, the methodology is generally robust, and the results are scientifically relevant and well placed within the context of previous Baltic Sea studies. The conclusions are largely supported by the data, although a stronger discussion of the uncertainties associated with the high percentage of gap-filled data and the resulting annual carbon balance estimates would further strengthen the paper. I believe the manuscript is suitable for publication after minor revisions focused on clarification of methodology, uncertainty assessment, and a few presentation improvements.
Other minor concern: Expand Figure 3 by including numerical importance values and uncertainty estimates, several statements should be softened from causal interpretations to predictor-based interpretations.