the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 19 May 2025
Cyclone enhances the contribution of oceanic dimethyl sulfide to the free troposphere over the Southern Ocean
Abstract. Under cold and clean atmospheric conditions, such as in free troposphere, oceanic dimethyl sulfide (DMS) was likely to form new particles. This is likely to happen over the Southern Ocean, where high DMS emissions occur along with frequent cyclones and storm activities which enhance vertical entrainment. Herein, the DMS contribution to free troposphere from the surface ocean was evaluated using the data collected from 34th Chinese Antarctic Research Expedition and Lana DMS emission climatology by running the Lagrangian particle dispersion model FLEXPART. Up to 13.1 % of the DMS was found to be transported to the free troposphere (altitudes above 2 km) from the surface ocean, which was enhanced by the cyclones. High DMS mixing ratios (> 100 pptv) were found surrounding the cyclones even at an altitude of 5 km. These results indicate that the significant DMS-derived new particles have probably occurred in high altitudes of the Southern Ocean.
- Preprint
(1072 KB) - Metadata XML
-
Supplement
(1277 KB) - BibTeX
- EndNote
Status: closed (peer review stopped)
- RC1: 'Comment on egusphere-2025-1622', Anonymous Referee #1, 09 Jun 2025
-
RC2: 'Comment on egusphere-2025-1622', Anonymous Referee #2, 10 Sep 2025
Review of MS egusphere-2025-1622
General comments
This article investigates dimethylsulfide (DMS) fluxes estimated from a Southern Ocean cruise and compares them with values from the Lana et al. (2011) climatology as inputs to the Lagrangian particle dispersion model FLEXPART. While the study addresses an important question about the links between ocean-derived DMS and vertical transport processes, several aspects require clarification and further evidence. In particular, the rationale for using the Lana et al. climatology, the specifics of the model setup, and the details of the study area are not sufficiently described. Observations of oceanic and atmospheric DMS are referred to throughout but not included. The analysis relies heavily on FLEXPART simulations to argue for a mechanistic link between ocean-derived DMS, low-pressure systems, and episodic vertical transport to the free troposphere. The authors propose that the coexistence of low-pressure systems and elevated surface atmospheric DMS concentrations suggests transport of DMS to higher altitudes, but this claim is not convincingly supported by the observational data or modelling results presented. It is also unclear which observational DMS mixing ratios are used in the model, and the interpretation of very low concentrations (< 0.5 ppt) appears overstated. At present, the modelling results are not strong enough to substantiate the central hypothesis. I do not find the evidence presented sufficient to support the hypothesis of significant cyclone driven DMS transport to the free troposphere. With clearer justification of the methods, stronger observational support, and a more cautious interpretation of the results, this work could make a valuable contribution. Therefore, I do not recommend the article for publication in its current form. Some specific comments are provided below.
Specific Comments
Introduction
Line 41: Suggest defining nM (nmol L-1)
Line 54: How are cyclones defined/identified by this study?
Line 58: Why use the old Lana et al. climatology and not the more recent and improved (including ~18-fold more data) Hulswar et al. climatology?
Method and material
Line 65: More details on the instrument, sampling setup and analysis are required.
- What depth/height are measurements made at?
- A brief description of the TOF and the sampling setup: how it works, pumps, flow rates, instrument precision and sensitivity, delay and response times, does the system reach equilibrium, and is that a requirement?
- What liquid/gas standards are used for calibration?
- When was calibration done?
- Did the data go through any quality control?
- Were values below LOD removed or replaced?
- How many / what percentage of values were below the LOD?
Results and discussion
Line 99: Where is the evidence for the clear association between DMS flux and ‘observed DMS hotspot (above 10 nM)’ shown? It would be useful to present the seawater and atmospheric concentrations, since they are heavily relied upon in this section.
Line 100: Again, where is this ‘negative DMS’ shown? And does this relate to seawater or atmospheric concentrations?
Line 102: Where is the Lana et al. estimated flux along the cruise track shown? How is their flux computed? Does it use the same gas transfer velocity as this study? Why does Lana et al. estimate a higher flux than this study?
Line 106: Does the ‘large-scale DMS hotspot’ refer to atmospheric or seawater DMS? This should be made more consistent and specific throughout. Again, why was this data (seawater and atmospheric DMS concentrations) not presented?
Line 108: In Zhang et al. (2020), atmospheric DMS was only measured at the surface. Is there any evidence to support the assertion of ‘vertical transport from the MBL to free troposphere’, as opposed to removal or horizontal transport of DMS close to the surface?
Line 112: Mixing heights ranging from 1-1.25 km (i.e., green colours in Fig. 1d) seem to cover quite a large area, not a small area as described.
Line 123: ‘mean modelling results’ – mean of what?
Line 124: Clarity required in terms of what modelling was done and how DMS was simulated. Some clarity needs to be provided regarding how the model has arrived at the concentrations it has.
- Is the cruise track considered a stationary point for the purposes of modelling?
- If so, which 1-degree grid cell was chosen as the point source? And why was this cell chosen?
- What DMS data was given to flexpart to simulate atmospheric DMS? Was an atmospheric concentration / surface flux value for the 1-degree grid cell given to calculate the dispersion? If a fixed concentration value was given to the model, please state it.
Line 127: As previously mentioned, where are the ‘high atmospheric DMS levels’ shown?
Line 128: ‘peaks of atmospheric DMS’ refer to simulated DMS and should be stated as such.
Line 129: What time steps are the 1 km, 3 km and 5 km panels at (Fig. 2a, b & c)? If these maps are representing simulated dispersion of surface atmospheric DMS, over what time period is this simulated transport occurring?
Line 130: Note that 0.5 ppt is extremely low atmospheric DMS concentrations (the scale of the colour bar in Fig. 2) so the idea of a 'strong influence' on vertical DMS transport seems to be an overstatement. Almost negligible concentrations of DMS are simulated to be transported to the free troposphere in Fig. 2. Concentrations discussed in this article are well below typical limits of detection for most atmospheric DMS sampling equipment. This point should be made clear, along with the significance of the findings.
Line 132: Again, it should be made clear that ‘atmospheric DMS peaks’ refer to simulated concentrations.
Line 141: Figures S1-S3 would suggest to me that atmospheric DMS near the surface is transported horizontally much more successfully than vertically in your simulations. It is also interesting that the concentrations in Figure S1 do not appear to diminish during your 2-week study period. DMS is understood to have a lifetime of up to only a few days, and you have specified a 2-day lifetime for modelling in the methods section. Can you explain how it can last weeks in your simulations?
Line 142: DMS mixing ratios are not ‘quite’ low, they are extremely low – 0.2 ppt is an almost negligible concentration.
Line 143: ‘Peaks’ should be ‘relative peaks’, since peaks of 0.2 ppt are close to zero.
Line 145: I would suggest that the 'peak' just disappears after 1 day, rather than moving eastward.
Line 145: Mixing ratios are not ‘relatively high’, they are extremely low.
Line 147: Mixing ratios are not ‘relatively high’, they are slightly elevated from zero (~0.04 ppt).
Line 151: Please define/specify the study area.
Line 154: Do you mean DMS mixing ratio at high altitudes over the SO is ‘only available’ over a certain time period or a certain region of the Southern Ocean? There is other Southern Ocean aircraft DMS data available (e.g., ATom, ORCAS). Also, please define ‘high altitude’. With regards to ‘near the oceans of New Zealand’, please be specific with the study area.
Line 156: Which HIPPO observations are included here? There were 5 flight campaigns 2009-2011. Do you select only the same months and/or hours of the day as your observations?
Line 158: It would be useful to show the mean line and shaded variability (presumably mean +/- 1SD?) for the HIPPO observations as well as the Lana simulation in Fig. S4. Please also properly label the Lana line (mean?) and shaded region (+/- 1SD?) in Fig. S4 to describe what they represent.
Line 166: Why is the 5km preferred over 1km or 3km in the main text?
Line 167: Why is this ‘as expected’?
Line 169: Which features of which panels in which figures specifically demonstrate this point?
Line 170: Suggest replacing ‘would’ with ‘could’.
Line 182: Define FT and H2SO4.
Line 206: Within the figure, I suspect ‘entertainment’ should read ‘entrainment’.
Summary
Line 238: Is ‘the vertical transportation of oceanic DMS is enhanced by the low-pressure systems’ quantified anywhere or just based on a qualitative (visual) assessment of the maps?
Citation: https://doi.org/10.5194/egusphere-2025-1622-RC2
Status: closed (peer review stopped)
-
RC1: 'Comment on egusphere-2025-1622', Anonymous Referee #1, 09 Jun 2025
Review of MS egusphere-2025-1622:
The manuscript by Zhang et al entitled “Cyclone enhances the contribution of oceanic dimethyl sulfide to the free troposphere over the Southern Ocean” deals with the measurement of oceanic and atmospheric dimethyl sulfide (DMS) during a cruise to the Southern Ocean which experienced low pressure systems. The authors have used a Lagrangian particle dispersion model (FLEXPART) to the observed data to generate a distribution pattern of DMS over 1, 3 and 5 kms. In addition, they also applied the model to the monthly climatology of DMS flux by Lana et al (2011) to generate similar altitude distribution pattern to compensate for the lack in their data as it was only along the ship track. Using the modelled data the authors want to drive the idea that cyclonic conditions will push oceanic DMS to much higher levels than under normal scenario. While the idea of the authors is commendable, the entire work is based on modelling, which is OK, but in the absence of any ground truthing and/or direct/indirect evidence of DMS or other derived products to higher levels under the influence of cyclone, the plot loses ground. Thus, I do not recommend publication of the manuscript in its present form. Please find below specific comments on the manuscript.
Introduction:
Line 54: The authors mention several cyclones, but in fact there were only 2 instances as shown in Fig. 1c and these seem to be low pressure systems. It is not clear whether they were cyclones. Satellite images as supplementary figures would have been provided.
Materials and methods:
Line 64: Details of underway shipboard measurements both in seawater and atmosphere is missing. What depth was the intake of seawater, similarly what height was the atmospheric air taken? A photo of the automated system as a supplementary figure would prove useul. How was the system calibrated?
Results and discussion:
Line 98: While the text mentions 27.5 µmol m-2 d-1, the scale in Fig. 1a is only up till 10 µmol m-2 d-1.
Line 102: Please give an explanation to why the DMS flux is lower than Lana climatology.
Line 127-128: The atmospheric DMS graph is not shown, thus its difficult to corroborate the sentence.
Citation: https://doi.org/10.5194/egusphere-2025-1622-RC1 -
RC2: 'Comment on egusphere-2025-1622', Anonymous Referee #2, 10 Sep 2025
Review of MS egusphere-2025-1622
General comments
This article investigates dimethylsulfide (DMS) fluxes estimated from a Southern Ocean cruise and compares them with values from the Lana et al. (2011) climatology as inputs to the Lagrangian particle dispersion model FLEXPART. While the study addresses an important question about the links between ocean-derived DMS and vertical transport processes, several aspects require clarification and further evidence. In particular, the rationale for using the Lana et al. climatology, the specifics of the model setup, and the details of the study area are not sufficiently described. Observations of oceanic and atmospheric DMS are referred to throughout but not included. The analysis relies heavily on FLEXPART simulations to argue for a mechanistic link between ocean-derived DMS, low-pressure systems, and episodic vertical transport to the free troposphere. The authors propose that the coexistence of low-pressure systems and elevated surface atmospheric DMS concentrations suggests transport of DMS to higher altitudes, but this claim is not convincingly supported by the observational data or modelling results presented. It is also unclear which observational DMS mixing ratios are used in the model, and the interpretation of very low concentrations (< 0.5 ppt) appears overstated. At present, the modelling results are not strong enough to substantiate the central hypothesis. I do not find the evidence presented sufficient to support the hypothesis of significant cyclone driven DMS transport to the free troposphere. With clearer justification of the methods, stronger observational support, and a more cautious interpretation of the results, this work could make a valuable contribution. Therefore, I do not recommend the article for publication in its current form. Some specific comments are provided below.
Specific Comments
Introduction
Line 41: Suggest defining nM (nmol L-1)
Line 54: How are cyclones defined/identified by this study?
Line 58: Why use the old Lana et al. climatology and not the more recent and improved (including ~18-fold more data) Hulswar et al. climatology?
Method and material
Line 65: More details on the instrument, sampling setup and analysis are required.
- What depth/height are measurements made at?
- A brief description of the TOF and the sampling setup: how it works, pumps, flow rates, instrument precision and sensitivity, delay and response times, does the system reach equilibrium, and is that a requirement?
- What liquid/gas standards are used for calibration?
- When was calibration done?
- Did the data go through any quality control?
- Were values below LOD removed or replaced?
- How many / what percentage of values were below the LOD?
Results and discussion
Line 99: Where is the evidence for the clear association between DMS flux and ‘observed DMS hotspot (above 10 nM)’ shown? It would be useful to present the seawater and atmospheric concentrations, since they are heavily relied upon in this section.
Line 100: Again, where is this ‘negative DMS’ shown? And does this relate to seawater or atmospheric concentrations?
Line 102: Where is the Lana et al. estimated flux along the cruise track shown? How is their flux computed? Does it use the same gas transfer velocity as this study? Why does Lana et al. estimate a higher flux than this study?
Line 106: Does the ‘large-scale DMS hotspot’ refer to atmospheric or seawater DMS? This should be made more consistent and specific throughout. Again, why was this data (seawater and atmospheric DMS concentrations) not presented?
Line 108: In Zhang et al. (2020), atmospheric DMS was only measured at the surface. Is there any evidence to support the assertion of ‘vertical transport from the MBL to free troposphere’, as opposed to removal or horizontal transport of DMS close to the surface?
Line 112: Mixing heights ranging from 1-1.25 km (i.e., green colours in Fig. 1d) seem to cover quite a large area, not a small area as described.
Line 123: ‘mean modelling results’ – mean of what?
Line 124: Clarity required in terms of what modelling was done and how DMS was simulated. Some clarity needs to be provided regarding how the model has arrived at the concentrations it has.
- Is the cruise track considered a stationary point for the purposes of modelling?
- If so, which 1-degree grid cell was chosen as the point source? And why was this cell chosen?
- What DMS data was given to flexpart to simulate atmospheric DMS? Was an atmospheric concentration / surface flux value for the 1-degree grid cell given to calculate the dispersion? If a fixed concentration value was given to the model, please state it.
Line 127: As previously mentioned, where are the ‘high atmospheric DMS levels’ shown?
Line 128: ‘peaks of atmospheric DMS’ refer to simulated DMS and should be stated as such.
Line 129: What time steps are the 1 km, 3 km and 5 km panels at (Fig. 2a, b & c)? If these maps are representing simulated dispersion of surface atmospheric DMS, over what time period is this simulated transport occurring?
Line 130: Note that 0.5 ppt is extremely low atmospheric DMS concentrations (the scale of the colour bar in Fig. 2) so the idea of a 'strong influence' on vertical DMS transport seems to be an overstatement. Almost negligible concentrations of DMS are simulated to be transported to the free troposphere in Fig. 2. Concentrations discussed in this article are well below typical limits of detection for most atmospheric DMS sampling equipment. This point should be made clear, along with the significance of the findings.
Line 132: Again, it should be made clear that ‘atmospheric DMS peaks’ refer to simulated concentrations.
Line 141: Figures S1-S3 would suggest to me that atmospheric DMS near the surface is transported horizontally much more successfully than vertically in your simulations. It is also interesting that the concentrations in Figure S1 do not appear to diminish during your 2-week study period. DMS is understood to have a lifetime of up to only a few days, and you have specified a 2-day lifetime for modelling in the methods section. Can you explain how it can last weeks in your simulations?
Line 142: DMS mixing ratios are not ‘quite’ low, they are extremely low – 0.2 ppt is an almost negligible concentration.
Line 143: ‘Peaks’ should be ‘relative peaks’, since peaks of 0.2 ppt are close to zero.
Line 145: I would suggest that the 'peak' just disappears after 1 day, rather than moving eastward.
Line 145: Mixing ratios are not ‘relatively high’, they are extremely low.
Line 147: Mixing ratios are not ‘relatively high’, they are slightly elevated from zero (~0.04 ppt).
Line 151: Please define/specify the study area.
Line 154: Do you mean DMS mixing ratio at high altitudes over the SO is ‘only available’ over a certain time period or a certain region of the Southern Ocean? There is other Southern Ocean aircraft DMS data available (e.g., ATom, ORCAS). Also, please define ‘high altitude’. With regards to ‘near the oceans of New Zealand’, please be specific with the study area.
Line 156: Which HIPPO observations are included here? There were 5 flight campaigns 2009-2011. Do you select only the same months and/or hours of the day as your observations?
Line 158: It would be useful to show the mean line and shaded variability (presumably mean +/- 1SD?) for the HIPPO observations as well as the Lana simulation in Fig. S4. Please also properly label the Lana line (mean?) and shaded region (+/- 1SD?) in Fig. S4 to describe what they represent.
Line 166: Why is the 5km preferred over 1km or 3km in the main text?
Line 167: Why is this ‘as expected’?
Line 169: Which features of which panels in which figures specifically demonstrate this point?
Line 170: Suggest replacing ‘would’ with ‘could’.
Line 182: Define FT and H2SO4.
Line 206: Within the figure, I suspect ‘entertainment’ should read ‘entrainment’.
Summary
Line 238: Is ‘the vertical transportation of oceanic DMS is enhanced by the low-pressure systems’ quantified anywhere or just based on a qualitative (visual) assessment of the maps?
Citation: https://doi.org/10.5194/egusphere-2025-1622-RC2
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 713 | 70 | 13 | 796 | 28 | 12 | 28 |
- HTML: 713
- PDF: 70
- XML: 13
- Total: 796
- Supplement: 28
- BibTeX: 12
- EndNote: 28
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Review of MS egusphere-2025-1622:
The manuscript by Zhang et al entitled “Cyclone enhances the contribution of oceanic dimethyl sulfide to the free troposphere over the Southern Ocean” deals with the measurement of oceanic and atmospheric dimethyl sulfide (DMS) during a cruise to the Southern Ocean which experienced low pressure systems. The authors have used a Lagrangian particle dispersion model (FLEXPART) to the observed data to generate a distribution pattern of DMS over 1, 3 and 5 kms. In addition, they also applied the model to the monthly climatology of DMS flux by Lana et al (2011) to generate similar altitude distribution pattern to compensate for the lack in their data as it was only along the ship track. Using the modelled data the authors want to drive the idea that cyclonic conditions will push oceanic DMS to much higher levels than under normal scenario. While the idea of the authors is commendable, the entire work is based on modelling, which is OK, but in the absence of any ground truthing and/or direct/indirect evidence of DMS or other derived products to higher levels under the influence of cyclone, the plot loses ground. Thus, I do not recommend publication of the manuscript in its present form. Please find below specific comments on the manuscript.
Introduction:
Line 54: The authors mention several cyclones, but in fact there were only 2 instances as shown in Fig. 1c and these seem to be low pressure systems. It is not clear whether they were cyclones. Satellite images as supplementary figures would have been provided.
Materials and methods:
Line 64: Details of underway shipboard measurements both in seawater and atmosphere is missing. What depth was the intake of seawater, similarly what height was the atmospheric air taken? A photo of the automated system as a supplementary figure would prove useul. How was the system calibrated?
Results and discussion:
Line 98: While the text mentions 27.5 µmol m-2 d-1, the scale in Fig. 1a is only up till 10 µmol m-2 d-1.
Line 102: Please give an explanation to why the DMS flux is lower than Lana climatology.
Line 127-128: The atmospheric DMS graph is not shown, thus its difficult to corroborate the sentence.