the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 25 Aug 2025
New insights in atmospheric methane variability in the Arctic by ship-borne measurements during MOSAiC
Abstract. The sparse network of Arctic land-based stations results in significant data gaps for atmospheric methane (CH4), particularly in sea-ice covered regions. Ship-based measurements can complement these data, improving understanding of regional and seasonal CH4 variability. This study presents continuous atmospheric ship-borne recordings of CH4 concentration and isotopic composition above the open ocean and sea-ice surface during Leg 4 (June–July 2020) and Leg 5 (August–September 2020) of the Multidisciplinary drifting Observatory for the Study of Arctic Climate (MOSAiC) expedition in the Central Arctic. Our measurements aim to enhance process understanding by identifying local emission sources and refining transport pathway and atmospheric mixing analysis. We compared three contamination-filtering methods and applied the Pollution Detection Algorithm to the raw data. Comparison with nearby land-based stations and their seasonal cycles suggests ship-borne data capture dynamic changes in CH4 sources, sinks, and transport processes, beyond seasonality. To unravel the underlying processes, we identified air mass transport pathways within the atmospheric boundary layer above the Arctic Ocean and their source areas using five-day backward trajectories from the LAGRANTO tool, based on ERA5 wind field data. Our analysis reveals that CH4 variability is driven by air masses predominantly influenced by open ocean and sea-ice-covered regions, with sea-ice dynamics imparting specific modifications along transport pathways. These findings underscore the importance of air mass transport and origin in shaping central Arctic CH4 variability. The study highlights the value of integrating ship-borne CH4 measurements with trajectory analysis to improve process-level understanding and support enhanced regional modeling.
- Preprint
(5626 KB) - Metadata XML
- BibTeX
- EndNote
Status: open (until 06 Oct 2025)
-
RC1: 'Comment on egusphere-2025-3778', Hélène Angot, 25 Aug 2025
reply
This manuscript presents continuous ship-borne measurements of atmospheric methane (CH4) concentration and isotopic composition during MOSAiC legs 4 and 5, complemented by daily backward trajectories to investigate transport pathways and source regions. The study is interesting and contributes valuable observations from an under-sampled region, where land-based stations are sparse and ship-based records can indeed help resolve spatial and temporal variability. The integration of CH4 isotopes with transport analysis has potential to advance our understanding of central Arctic CH4 dynamics.
However, I see a major limitation: the study does not sufficiently connect with, or make use of, the broader suite of datasets that were collected during MOSAiC. This weakens the contextualization and originality of the work. In particular:
1. Existing CH4 datasets from MOSAiC
The authors repeatedly highlight the “need for increased spatiotemporal data resolution, including year-round measurements”. These datasets already exist. In Angot et al. (2022), continuous year-round atmospheric CH4 measurements from MOSAiC are reported and made publicly available via PANGAEA (Angot et al., 2022). Acknowledging, referencing, and ideally comparing with these existing records would considerably strengthen the manuscript and properly situate this study within the broader MOSAiC framework.
2. Air mass transport and trajectory analysis
The manuscript relies on daily back trajectories from LAGRANTO. While this provides some insight, a much richer dataset is available: high-resolution FLEXPART footprints and multiple trajectories per day, including emission sensitivities over land, ocean, and sea ice (see https://webdata.wolke.img.univie.ac.at/mosaic/mosaic.html). Incorporating or at least discussing this dataset would provide a much more detailed and robust analysis of air mass origins and transport, compared to the current daily trajectory approach.
Additional comments
3. Calibration of the instrument
The Methods section does not describe the calibration procedures used during the expedition. Furthermore, the authors state that “mean CH4 concentrations were approximately 5% lower than the simultaneous measured mean levels at the nearby land-based stations”. This is interpreted as a potential dilution of the signal during long-range transport over sea ice. However, the reported concentrations appear low compared to the merged MOSAiC dataset (Angot et al., 2022), which is traceable to the WMO-X2004A calibration scale. A direct intercomparison would help clarify whether this difference reflects calibration issues, transport effects, or other causes.
4. Pollution detection algorithm
The authors note that three contamination-filtering approaches were tested, and that the Beck et al. (2022) Pollution Detection Algorithm was ultimately chosen. Since pollution filtering is a critical and often challenging step, I would encourage the authors to show these results, at least in the Supplementary Information. For example, comparing datasets flagged by different methods would provide a useful sanity check. It would also be valuable to compare the final QC’d dataset with that of Angot et al. (2022) to assess consistency between independently collected and cleaned datasets.
5. Seasonal cycle and short-term variability
The manuscript reports a “most pronounced concentration increase during August”, relatively stable concentrations near the North Pole (leg 5), and higher variability close to Svalbard (leg 4). How does this compare with the year-round dataset reported by Angot et al. (2022) (see attached Figure showing monthly distributions)? For example, that dataset shows a large spike in early January (contextualized in Boyer et al. (2023)), and higher variability in July–September compared to the autumn and spring periods. A discussion linking the ship-borne observations to this broader seasonal context would be highly informative.
6. Isotopic composition
The isotopic data are of great interest but appear somewhat underutilized. I would have expected a more detailed discussion of possible source contributions, such as Siberian anthropogenic point sources or wildfire plumes (e.g., the July 27–29 event). Further interpretation of the isotopic signatures could add significant value to the manuscript.
Recommendation: This study provides valuable data, but in its current form it underutilizes the wealth of complementary MOSAiC datasets. A revision that (i) acknowledges and integrates year-round CH4 observations, (ii) considers high-resolution transport datasets, and (iii) provides more detailed discussion of calibration, pollution filtering, and isotopic signatures would substantially improve the robustness, originality, and impact of the work.
References
Angot, H., Blomquist, B., Howard, D., Archer, S., Bariteau, L., Beck, I., Helmig, D., et al. (2022) ‘Methane dry air mole fractions measured during MOSAiC 2019/2020 (merged dataset)’. PANGAEA. Available at: https://doi.pangaea.de/10.1594/PANGAEA.944291.Angot, H., Blomquist, B., Howard, D., Archer, S., Bariteau, L., Beck, I., Boyer, M., et al. (2022) ‘Year-round trace gas measurements in the central Arctic during the MOSAiC expedition’, Scientific Data, 9(1), p. 723. Available at: https://doi.org/10.1038/s41597-022-01769-6.
Beck, I. et al. (2022) ‘Automated identification of local contamination in remote atmospheric composition time series’, Atmospheric Measurement Techniques, 15(14), pp. 4195–4224. Available at: https://doi.org/10.5194/amt-15-4195-2022.
Boyer, M. et al. (2023) ‘A full year of aerosol size distribution data from the central Arctic under an extreme positive Arctic Oscillation: insights from the Multidisciplinary drifting Observatory for the Study of Arctic Climate (MOSAiC) expedition’, Atmospheric Chemistry and Physics, 23(1), pp. 389–415. Available at: https://doi.org/10.5194/acp-23-389-2023.
-
RC2: 'Comment on egusphere-2025-3778', Anonymous Referee #2, 02 Sep 2025
reply
This is a well-written paper on a potentially informative and important topic, especially so given the limited availability of methane time-series data from the Arctic Ocean. This manuscript relies substantially on the interpretation of airmass back trajectories and my only significant concern deals with how trajectories are used in this analysis.
According to the description in Section 2.3, a cluster of back trajectories is computed each day from an area within 30 km of the ship’s position, on a 5 km grid. The maps in Figure 2 show paths for back-trajectory clusters on two dates, illustrating what the authors describe as ‘over ice’ and ‘open ocean - terrestrial’ characteristics. My main question is how the authors account for altitude variability in these computed trajectories, as this is not shown in a Figure or discussed in Section 2.3.
I believe these are isentropic back trajectories, and pressure altitude along the time steps should be one of the outputs. The airmass will only be in contact with surface sources of CH4 if it is at the surface, within the boundary layer, which can be very shallow over the ice in the Arctic summer and is capped by a typically strong inversion. Over land or open ocean the BL can be much deeper. An airmass back trajectory in the free troposphere will not be influenced by surface emissions.
So, how exactly do the authors classify the significance of surface types and emissions over the back trajectory? Do they exclude portions of the back trajectory that are above the BL? If so, are you using estimates of BL depth from ERA5? From the text in Section 2.3 it sounds like the surface type under the entire length of the 5-day trajectory is used to derive the classification and to identify potential sources of observed variability in the CH4 concentration. But this cannot be a correct approach when significant portions of the trajectory are above the BL, and thus out of contact with the surface. We need more explanation of the methodology here to understand the analysis.
One final comment concerns the comparison with Arctic time series stations operated by NOAA/GML, AGAGE, etc. It was helpful and informative to show some of this data on Figure 3. The authors should be aware that NOAA/GML also took weekly flask samples at the Polarstern site during MOSAiC and these have been analyzed in their Boulder lab following the same rigorous protocols as other long-term GML monitoring sites. It would be interesting to also see this data presented with your time series.
Citation: https://doi.org/10.5194/egusphere-2025-3778-RC2
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 1,528 | 23 | 3 | 1,554 | 10 | 15 |
- HTML: 1,528
- PDF: 23
- XML: 3
- Total: 1,554
- BibTeX: 10
- EndNote: 15
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1