the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 12 Nov 2025
Assessing methane emissions from an offshore marine aggregate extraction site near Sylt, Eastern North Sea
Abstract. Coastal regions are estimated to contribute with up to 1 % to the global atmospheric methane (CH4) budget. However, these emissions remain highly uncertain due to strong spatial and temporal variability. Anthropogenic activities in coastal waters, such as sand mining and beach nourishment, have not yet been investigated or included in the calculation of the resulting greenhouse gas emissions. Since 1984, regular beach nourishment has been carried out every summer on the west coast of the island of Sylt (North Sea, Germany). During this process, dredging vessels extract a mixture of sand and water from the seabed of a spatially confined area (Westerland II), situated approximately 8 km off the coast and deposit the material on the western beach and foreshore of Sylt. High resolution measurements of CH4 in ambient air at the coastal atmospheric station Westerland (Sylt, Germany) show CH4 spikes of up to 400 ppb above background concentrations. These spikes occurred mainly during summer season, during low tide and under westerly wind conditions (from the sand dredging area).
To investigate the origin of the observed atmospheric CH4 spikes, combined in-situ measurements of dissolved and atmospheric CH4 together with water sampling, were performed on board the research vessel RV Mya II (AWI) along a coastal transect from Sylt to the Westerland II dredging site. In the vicinity of the dredging site, elevated CH4 concentrations were detected both in ambient air (400–500 ppb above background reaching up to 2450 ppb) and in surface and bottom waters (69–90 nmol L-1). This resulted in a mean diffusive flux of 45 ± 47 µmol m-2 d-1 with a maximum value of 340 µmol m-2 d-1 in the dredging area, in contrast to a diffusive flux of 3.0 ± 3.6 µmol m-2 d-1 during the transit. Our observations demonstrate significantly higher CH4 concentrations and fluxes in both the atmosphere and the water column above the sand dredging site, compared to the areas outside the dredging activities. Furthermore, the isotopic composition of dissolved CH4 within the dredging site characterized by more negative stable carbon and hydrogen isotope values, point to a microbial source of the excess CH4.
-
Notice on discussion status
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
-
Preprint
(1753 KB)
-
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(1753 KB) - Metadata XML
- BibTeX
- EndNote
- Final revised paper
Journal article(s) based on this preprint
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2025-4850', Anonymous Referee #1, 19 Dec 2025
-
AC1: 'Reply on RC1', Martina Schmidt, 11 Mar 2026
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-4850/egusphere-2025-4850-AC1-supplement.pdf
-
AC1: 'Reply on RC1', Martina Schmidt, 11 Mar 2026
-
RC2: 'Comment on egusphere-2025-4850', Anonymous Referee #2, 18 Feb 2026
Review of “Assessing methane emissions from an offshore marine aggregate extraction site near Sylt, Eastern North Sea” by Schmidt et al.,
The paper by Schmidt et al. describes the result of a 1-day field study to identify the source of methane from a sand extraction site offshore Sylt, which had been recognized based on continuous measurements of the atmospheric mole fraction of the ICOS station Westerland. While the findings are interesting and surely deserve publication, and while it is clear that several questions cannot be answered by the result of a one day survey, there are some points that should be clarified or exploited a little bit more before final publication. This bears also the potential to increase the impact of the study. The most important suggestion is to expand the exploitation of the atmospheric data from Westerland, which could help to add substantiate some statements which are so far rather unsupported.
I start to address the major points, followed by some minor mostly editorial aspects.
Major:
The authors should expand the information on the magnitude and distribution of sand extraction in the North Sea/European/Global context in the introduction. They do so later in the discussion, also reasoning why extrapolation of results is difficult, but should put some information in the introduction. That would leave room in the discussion to be even a little bit more quantitative, potentially introduce a European wide map etc. In that regard, the short statement on the European perspective in Lines 99-100 should be moved out of the site description, where it does not belong.
Methodology:
Lines 119 to 120: Instrumentation of the meteorological measurements should be given.
Line 137: while it is stated how time lapse was corrected for the atm. measurements, the delay and info on flow rate / potential contamination in the pump/bucket, water supply should be briefly addressed
Line 147: it is mentioned that the ASE exchange is calculated after Wanninkhof 2009, , while line 154 and 155 clearly state that Nitghingale 2000 was used. Was something else extracted from Wanninkhof 2009; or is this just an erratum?
Line 179: it is stated that only one sample was measured for surface waters, but in line 167, it is stated that there were 3 samples taken for each surface point. Likely, the authors measured all three samples per site. So for the surface, you measured 3x one sample and for the deep water , you took 3 measurements from one sample? Please clarify.
Missing information in methods:
It is mentioned that two air-inlets, one at 8 m and one handheld inlet for bottom near atmospheric sampling, were available, to be selected by a 3 port valve. But it is not clear which inlet was used when. So far, there is only indication for one transect sampled twice, one time at 8 m and one time handheld near the surface. Was that the only occasion the sea surface near sampling was performed. Please specify.
In line 366 (discussion), it is mentioned that earlier mobile air sampling was used to come to the conclustion that the methane source detected at ICOS station Westerland was likely offshore and likely in direction of the extraction site. However, this part is not mentioned at all in the method part. Please add.
Results :
It is not clear for me why the atmospheric duration of the record chosen to demonstrate the detection of the spikes in CH4 is so short (Fig 3). Apparently, the authors picked the same period in 2022 as is discussed in 2023 in connection with the campaign. However, data from station Westerland are available for a long time (as the dredging activity). Moreover, the high frequency data should be available at least since 2021, though the effect might be traceable also in the 1 min data set obtained before. The value of the atmospheric record over a longer time period is that several assumptions and statements could be proven by looking at the atmospheric data for a longer time period , together with wind re-analysis data (or local wind data from the station).
- The statement that this phenomenon occur mostly in summer; more specifically, that it occurs mostly while active dredging occurs.
- The statement that the emissions occur mostly during low tide, which cannot be judged based on these few days that are displayed currently in Fig. 3.
- This analysis could also further narrow down the wind direction which is required for the emissions to be visible in the station’s record. This could shed some light on the question whether the weaker signal in 2023 and some of the variability observed in the strength is a function of the wind direction rather than the source strength.
Apparently, this work was at least partly done for the year 2022, as indicated in the lines 247-254. I think this part should be moved up, before the results for 2023 (so avoiding the 2022-2023-2022 jump), and substantiated by Figures /Tables. Currently these very important statements are based on data not shown.
It would be also interesting whether there are times in the dredging season when no ship is operating. So the question on whether the emissions are connected to the active action of extraction or to the characteristics of the young seafloor depressions caused by the extraction could be narrowed down further. This is particular important as direct linkage to the extraction action, which is somehow implied by the field data (highest emissions in direct vicinity to the extraction activity and related to a plume of suspended material) is difficult to be linked to a tidal control (assuming there is no reason to believe that extraction activities are mostly executed during low tide, something which would be important to know).
Figure 5 and Lines 273-277
If I understood it correctly, one transect near the ship was measured twice, one time with the 8m inlet and one time with the near—surface handheld inlet, and the considerable difference of 63 ppb (mean), the authors infer that there is a local source, leading to the near surface gradient. I strongly suggest to show these data separately, either in the text or in the appendix, as two distinguishable data sets against latitude, longitude, or distance, and indicate the position in Figure 5. The presentation in Fig 5 is not suitable to display the finding.
Line 293 to 295: The sentence starting with “The comparison between …”. I am dismayed, but I cannot understand the argument given in this sentence. Maybe rephrase?
Line 308-309: what do you mean by “similar range” here. As you state correctly, the Westerland II values are clearly more positive (or less negative).
Line 346 and Fig 10b. Given the limited range of data and the relatively large error range of the data, I found the error on the intercept (66 ± 6) astonishingly small. And I was astonished that, while you apparently used a York fit because of uncertainties of the data in both dimensions, the data displayed have only an error indicated for the Delta value. Please check.
Discussion
Lines 366 to 368. As mentioned further up, this comes “out of the blue” as these measurements were not introduced or specified in the method section.
Line 378: “Therefore we did not expect …”. Please extend a little on the rationale. One could also argue, given the shallow water depth, that the storm induced resuspension leading to enhanced concentration. Apparently not the case, but the rationale for the statement is not explained.
Lines 386-389: Using READY to estimate the relation between offshore source strength and signal strength at the Westerland station adds value to the manuscript. However, it is not fully exploited. To what extend could larger peaks also result from a better matching wind direction or differences in flow regime? It would also be interesting to do a run for some of the very strong peaks observed in the 2022 data, which appear substantially larger than 15 ppb. As you ran the model already, adding this information might be easily done.
Line 390: as mentioned, I think something on the extend of dredging for marine aggregates should be already mentioned in the introduction, and you might consider to extend a map here, at least in the appendix.
Lines 468-470. The attribution of the concentration and isotopic findings at station #4 to oxidation is questionable, given the very short residence time of the water in the suggested methane source area. Turnover rates at methane concentration levels as reported in this study are likely weeks to months (e.g. Nauw et al., 2015). Displacement of water, i.e. hydrographic reasons, appear more likely.
Last sentence of conclusion: While the findings of this work are really novel and exiting, before calling for long-term observation and mitigation strategies, an assessment of the importance of the process in comparison to other sources would be needed.
Minor:
Line 30: … dredging site COMMA characterized …
Line 59: ..inputs from riverine => from rivers
Line 72: approximate volume of more than .. change to approximate volume of 57 million …. (the No is later specified to 56.9)
Line 91: … depth of 20m below seafloor => 20 m deep (you characterize the depression relative to the surrounding.
Line 92: The infill … ; sentence has to many “and”; maybe split to two sentences?
Fig 1: please give source of bathymetric data
Line 357: … (wind from the west) requires => calls for
Line 370: “Circling around the vessel resulted in increased atmospheric CH4 concentrations” Please rephrase – the increased concentrations are not a result of the movement of the Mya.
Line 378: “In addition, storm Poly, passed …” ; Remove the commas
Lines 406-407: Dissolved CH4 concentrations have been measured in the area south of our study area, but with the same method. => … study area, using the same method; I cannot see the contradiction indicated by the word “but”.
Line 479: remove one “the ship campaign”
Line 480 : none AT all
Line 482-483: “Such atmospheric measurements …”. I do not understand the argument. How can you calculate the emissions just from atm. concentration measurements? And do you refer to the atm. measurements at station Westerland or at the extraction site.
Citation: https://doi.org/10.5194/egusphere-2025-4850-RC2 -
AC2: 'Reply on RC2', Martina Schmidt, 11 Mar 2026
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-4850/egusphere-2025-4850-AC2-supplement.pdf
Peer review completion
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2025-4850', Anonymous Referee #1, 19 Dec 2025
In "Assessing methane emissions from an offshore marine aggregate extraction site near Sylt, Eastern North Sea", the authors present the results of a one-day ship campaign aimed at confirming that the excess CH4 sometimes measured in summer at the ICOS site Westerland are coming from the sand extraction site Westerland II. To do so, they combine air and surface and bottom water sampling and measure CH4 but its isotopes as well for a few samples.
The manuscript is mostly clear and well written and should be published subject to minor revisions.
In general, the manuscript should be reread to make sure that people from both communities (atmosphere and ocean) fully understand each other parts and that all measurements are well presented. Below, I underline the points that I think would improve the manuscript.
p4 l143 "To convert the relative concentrations (ppm) reported by the GGA to absolute concentrations (nmol L-1), discrete water samples were collected at least once per hour (Bussmann et al., 2024)." can you please detail a bit this part? Why the temporality is needed to get the conversion from ppm to nmol/L?
p5 l149 what value is used for ceq?
p5 l159 Can you shortly define the Schmidt number?
p5 l165 please define CTD (also in Figure 1)
In 2.4, can you add the actual number of samples taken for the headspace measurements and for the isotopes? Listed only in 3.4, you say you have 4 stations but 11 samples. please clarify.
In Figure 5, can you add the location of the atmospheric isotopic samples ? Also, check the legend , 1a and 1b are only blue not black and blue (or clarify)
In Figure 7, can you add the location of the 11 isotopic samples (1 to 11)?
p12 3.6 you only describe the location of 4 bags, what about the others?
Minor corrections:
p1 l23 please define AWI
p2 l48 replace gromers by groomers
p4 l118 replace ICOS-Cal by ICOS-FCL
p14 l407 "Dissolved CH4 concentrations have been measured in the area south of our study area, but with the same method." Should it no read "but with a different method"?
p16 l480 replace non et all by none at all
p18 l523 a parenthesis is missing
Citation: https://doi.org/10.5194/egusphere-2025-4850-RC1 -
AC1: 'Reply on RC1', Martina Schmidt, 11 Mar 2026
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-4850/egusphere-2025-4850-AC1-supplement.pdf
-
AC1: 'Reply on RC1', Martina Schmidt, 11 Mar 2026
-
RC2: 'Comment on egusphere-2025-4850', Anonymous Referee #2, 18 Feb 2026
Review of “Assessing methane emissions from an offshore marine aggregate extraction site near Sylt, Eastern North Sea” by Schmidt et al.,
The paper by Schmidt et al. describes the result of a 1-day field study to identify the source of methane from a sand extraction site offshore Sylt, which had been recognized based on continuous measurements of the atmospheric mole fraction of the ICOS station Westerland. While the findings are interesting and surely deserve publication, and while it is clear that several questions cannot be answered by the result of a one day survey, there are some points that should be clarified or exploited a little bit more before final publication. This bears also the potential to increase the impact of the study. The most important suggestion is to expand the exploitation of the atmospheric data from Westerland, which could help to add substantiate some statements which are so far rather unsupported.
I start to address the major points, followed by some minor mostly editorial aspects.
Major:
The authors should expand the information on the magnitude and distribution of sand extraction in the North Sea/European/Global context in the introduction. They do so later in the discussion, also reasoning why extrapolation of results is difficult, but should put some information in the introduction. That would leave room in the discussion to be even a little bit more quantitative, potentially introduce a European wide map etc. In that regard, the short statement on the European perspective in Lines 99-100 should be moved out of the site description, where it does not belong.
Methodology:
Lines 119 to 120: Instrumentation of the meteorological measurements should be given.
Line 137: while it is stated how time lapse was corrected for the atm. measurements, the delay and info on flow rate / potential contamination in the pump/bucket, water supply should be briefly addressed
Line 147: it is mentioned that the ASE exchange is calculated after Wanninkhof 2009, , while line 154 and 155 clearly state that Nitghingale 2000 was used. Was something else extracted from Wanninkhof 2009; or is this just an erratum?
Line 179: it is stated that only one sample was measured for surface waters, but in line 167, it is stated that there were 3 samples taken for each surface point. Likely, the authors measured all three samples per site. So for the surface, you measured 3x one sample and for the deep water , you took 3 measurements from one sample? Please clarify.
Missing information in methods:
It is mentioned that two air-inlets, one at 8 m and one handheld inlet for bottom near atmospheric sampling, were available, to be selected by a 3 port valve. But it is not clear which inlet was used when. So far, there is only indication for one transect sampled twice, one time at 8 m and one time handheld near the surface. Was that the only occasion the sea surface near sampling was performed. Please specify.
In line 366 (discussion), it is mentioned that earlier mobile air sampling was used to come to the conclustion that the methane source detected at ICOS station Westerland was likely offshore and likely in direction of the extraction site. However, this part is not mentioned at all in the method part. Please add.
Results :
It is not clear for me why the atmospheric duration of the record chosen to demonstrate the detection of the spikes in CH4 is so short (Fig 3). Apparently, the authors picked the same period in 2022 as is discussed in 2023 in connection with the campaign. However, data from station Westerland are available for a long time (as the dredging activity). Moreover, the high frequency data should be available at least since 2021, though the effect might be traceable also in the 1 min data set obtained before. The value of the atmospheric record over a longer time period is that several assumptions and statements could be proven by looking at the atmospheric data for a longer time period , together with wind re-analysis data (or local wind data from the station).
- The statement that this phenomenon occur mostly in summer; more specifically, that it occurs mostly while active dredging occurs.
- The statement that the emissions occur mostly during low tide, which cannot be judged based on these few days that are displayed currently in Fig. 3.
- This analysis could also further narrow down the wind direction which is required for the emissions to be visible in the station’s record. This could shed some light on the question whether the weaker signal in 2023 and some of the variability observed in the strength is a function of the wind direction rather than the source strength.
Apparently, this work was at least partly done for the year 2022, as indicated in the lines 247-254. I think this part should be moved up, before the results for 2023 (so avoiding the 2022-2023-2022 jump), and substantiated by Figures /Tables. Currently these very important statements are based on data not shown.
It would be also interesting whether there are times in the dredging season when no ship is operating. So the question on whether the emissions are connected to the active action of extraction or to the characteristics of the young seafloor depressions caused by the extraction could be narrowed down further. This is particular important as direct linkage to the extraction action, which is somehow implied by the field data (highest emissions in direct vicinity to the extraction activity and related to a plume of suspended material) is difficult to be linked to a tidal control (assuming there is no reason to believe that extraction activities are mostly executed during low tide, something which would be important to know).
Figure 5 and Lines 273-277
If I understood it correctly, one transect near the ship was measured twice, one time with the 8m inlet and one time with the near—surface handheld inlet, and the considerable difference of 63 ppb (mean), the authors infer that there is a local source, leading to the near surface gradient. I strongly suggest to show these data separately, either in the text or in the appendix, as two distinguishable data sets against latitude, longitude, or distance, and indicate the position in Figure 5. The presentation in Fig 5 is not suitable to display the finding.
Line 293 to 295: The sentence starting with “The comparison between …”. I am dismayed, but I cannot understand the argument given in this sentence. Maybe rephrase?
Line 308-309: what do you mean by “similar range” here. As you state correctly, the Westerland II values are clearly more positive (or less negative).
Line 346 and Fig 10b. Given the limited range of data and the relatively large error range of the data, I found the error on the intercept (66 ± 6) astonishingly small. And I was astonished that, while you apparently used a York fit because of uncertainties of the data in both dimensions, the data displayed have only an error indicated for the Delta value. Please check.
Discussion
Lines 366 to 368. As mentioned further up, this comes “out of the blue” as these measurements were not introduced or specified in the method section.
Line 378: “Therefore we did not expect …”. Please extend a little on the rationale. One could also argue, given the shallow water depth, that the storm induced resuspension leading to enhanced concentration. Apparently not the case, but the rationale for the statement is not explained.
Lines 386-389: Using READY to estimate the relation between offshore source strength and signal strength at the Westerland station adds value to the manuscript. However, it is not fully exploited. To what extend could larger peaks also result from a better matching wind direction or differences in flow regime? It would also be interesting to do a run for some of the very strong peaks observed in the 2022 data, which appear substantially larger than 15 ppb. As you ran the model already, adding this information might be easily done.
Line 390: as mentioned, I think something on the extend of dredging for marine aggregates should be already mentioned in the introduction, and you might consider to extend a map here, at least in the appendix.
Lines 468-470. The attribution of the concentration and isotopic findings at station #4 to oxidation is questionable, given the very short residence time of the water in the suggested methane source area. Turnover rates at methane concentration levels as reported in this study are likely weeks to months (e.g. Nauw et al., 2015). Displacement of water, i.e. hydrographic reasons, appear more likely.
Last sentence of conclusion: While the findings of this work are really novel and exiting, before calling for long-term observation and mitigation strategies, an assessment of the importance of the process in comparison to other sources would be needed.
Minor:
Line 30: … dredging site COMMA characterized …
Line 59: ..inputs from riverine => from rivers
Line 72: approximate volume of more than .. change to approximate volume of 57 million …. (the No is later specified to 56.9)
Line 91: … depth of 20m below seafloor => 20 m deep (you characterize the depression relative to the surrounding.
Line 92: The infill … ; sentence has to many “and”; maybe split to two sentences?
Fig 1: please give source of bathymetric data
Line 357: … (wind from the west) requires => calls for
Line 370: “Circling around the vessel resulted in increased atmospheric CH4 concentrations” Please rephrase – the increased concentrations are not a result of the movement of the Mya.
Line 378: “In addition, storm Poly, passed …” ; Remove the commas
Lines 406-407: Dissolved CH4 concentrations have been measured in the area south of our study area, but with the same method. => … study area, using the same method; I cannot see the contradiction indicated by the word “but”.
Line 479: remove one “the ship campaign”
Line 480 : none AT all
Line 482-483: “Such atmospheric measurements …”. I do not understand the argument. How can you calculate the emissions just from atm. concentration measurements? And do you refer to the atm. measurements at station Westerland or at the extraction site.
Citation: https://doi.org/10.5194/egusphere-2025-4850-RC2 -
AC2: 'Reply on RC2', Martina Schmidt, 11 Mar 2026
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-4850/egusphere-2025-4850-AC2-supplement.pdf
Peer review completion
Journal article(s) based on this preprint
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 2,086 | 824 | 170 | 3,080 | 144 | 174 |
- HTML: 2,086
- PDF: 824
- XML: 170
- Total: 3,080
- BibTeX: 144
- EndNote: 174
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Martina Schmidt
Ingeborg Bussmann
Sarah Johanna Ernestina Reith
Annika Palzer
Cedric Couret
Frank Keppler
Daniela Polag
Lasse Sander
Moritz Schroll
Julia B. Wietzel
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(1753 KB) - Metadata XML
In "Assessing methane emissions from an offshore marine aggregate extraction site near Sylt, Eastern North Sea", the authors present the results of a one-day ship campaign aimed at confirming that the excess CH4 sometimes measured in summer at the ICOS site Westerland are coming from the sand extraction site Westerland II. To do so, they combine air and surface and bottom water sampling and measure CH4 but its isotopes as well for a few samples.
The manuscript is mostly clear and well written and should be published subject to minor revisions.
In general, the manuscript should be reread to make sure that people from both communities (atmosphere and ocean) fully understand each other parts and that all measurements are well presented. Below, I underline the points that I think would improve the manuscript.
p4 l143 "To convert the relative concentrations (ppm) reported by the GGA to absolute concentrations (nmol L-1), discrete water samples were collected at least once per hour (Bussmann et al., 2024)." can you please detail a bit this part? Why the temporality is needed to get the conversion from ppm to nmol/L?
p5 l149 what value is used for ceq?
p5 l159 Can you shortly define the Schmidt number?
p5 l165 please define CTD (also in Figure 1)
In 2.4, can you add the actual number of samples taken for the headspace measurements and for the isotopes? Listed only in 3.4, you say you have 4 stations but 11 samples. please clarify.
In Figure 5, can you add the location of the atmospheric isotopic samples ? Also, check the legend , 1a and 1b are only blue not black and blue (or clarify)
In Figure 7, can you add the location of the 11 isotopic samples (1 to 11)?
p12 3.6 you only describe the location of 4 bags, what about the others?
Minor corrections:
p1 l23 please define AWI
p2 l48 replace gromers by groomers
p4 l118 replace ICOS-Cal by ICOS-FCL
p14 l407 "Dissolved CH4 concentrations have been measured in the area south of our study area, but with the same method." Should it no read "but with a different method"?
p16 l480 replace non et all by none at all
p18 l523 a parenthesis is missing