the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 30 Aug 2021
Southern Ocean latitudinal gradients of cloud condensation nuclei
Ruhi S. Humphries
Melita D. Keywood
Sean Gribben
Ian M. McRobert
Jason P. Ward
Paul Selleck
Sally Taylor
James Harnwell
Connor Flynn
Gourihar R. Kulkarni
Gerald G. Mace
Alain Protat
Simon P. Alexander
Greg McFarquhar
Download
- Final revised paper (published on 30 Aug 2021)
- Preprint (discussion started on 27 Jan 2021)
Interactive discussion
Status: closed
-
RC1: 'Comment on acp-2020-1246', Anonymous Referee #1, 11 Mar 2021
This paper by Humphries et al. "Southern Ocean latitudinal gradients of Cloud Condensation Nuclei" presents important new observations of CCN for the sparsely measured region and describes valuable datasets. The authors have prepared a clear, well written manuscript. Please consider expanding the discussion in a few places, where it can add value. I suggest publication with minor revisions.Line 18: Long-term measurements at land-based research stations surrounding the Southern Ocean were found to be good representations at their respective latitudes. The long term stations should be regarded as the point for comparison with the much smaller ship-based dataset, e.g. "The ship-based measurements were in good agreement with the long-term measurements at theri respective latitudes. The voyage data are only snapshots in time. This is acknowledged towards the end of the discussion (line 435).Line 38 and following: "in-situ observations in this region rare, and until recently, only a handful of aerosol measurements" Agreed, and for completeness, it would be good to include reference to earlier work on latitudinal gradients of CCN, in particular Bigg, 1990.Line 195 and following:The authors have a thorough approach to treatment of engine exhaust contamination. Has the possibility of other lesser aerosol contamination been evaluated e.g. vent emissions from the cooking in the galley or from indoor air?Line 315: also supported by the analyses of Vallina et al., 2006 In addition Fossum et al 2020 suggest that lower sea salt may also mean that a suppression mechanism for sulfate aerosol activation is removed at high latitudeLine 395 and following: "despite being very difficult to identify in variables by which the front is defined (i.e. meteorological)". This wording unclear, please be more specific to improve and clarify. Is this saying that there is not a sharp boundary in winds at the polar front (around 60°S between the polar easterlies in the Polar Cell to the south and westerlies to the north compared to the contrast observed in the Atmospheric Compositional Front term that you introduce? I question if this term adds significantly to the understanding or whether it is an additional way of defining the Polar atmospheric Front. I can see that the the Antarctic Sea Ice Atmospheric Compositional Zone (ASIACZ) will have unique properties and it is good to highlight this where low aerosol surface area air descending off the polar plateau in the Polar Cell aided with katabatic flow will meet the coastal marine air where there is significant production of biogenic gases and potential for new particle formation. I note the phenomenon was commented on in earlier work by Wylie et al., 1993 You also raise an important point that this area will not necessarily be well sampled by Antarctic coastal stations especially if they are sitting in areas dominated by surface outflows. Further, Bigg 1990 found a peak in CCN around 60°S which coincided with the northern limit of sea-ice with the suggestion that the a peak in surface aerosol precursors could coincide with the edge of the seasonal sea-ice.Technical/Minor revisions:**************************Line 212: More precise to change "standard marine grade fuel oil" to "residual (heavy) fuel oil used by many vessels". https://www.sciencedirect.com/topics/earth-and-planetary-sciences/bunker-fuelLine 251: Can a bit more explanation be given on significance of the trajectory kde bandwidth smoothing using Scott's factor?Line 282: Change "southern ocean" to "Southern Ocean"Line 299 and 389: Re "step change - a trend in concentrations is apparent" but Fig 2, A2, A3 do not seem to suggest a "step change" With Ref to Fig 2, it appears that the largest difference is in CCN05. compared to other variables. Can comment be made on this? Further, the step change is most apparent in CCNx/CN ratio in Fig 3. It is useful to discuss this finding against the large scale "averaged CCN04/CN ratio of ~0.4 reported by Andreae 2009 across a range of environments, indicating an extreme in ratio with polar air.Line300 "in the CCN/CN ratios (Figures 2, A3, A4, A5, A6)," These seem to be presented in Figures 3, A6 and A7)Line 400: Change "is not captured" to "are not captured"Line 412: change "motivations" to "motivation"Figure A7: Rightmost label on abscissa is missing. This figure is both MARCUS and CAPRICORN2. Why is the 65-70 category as shown in Figure 2 not included?Refs:Andreae, M.O. (2009) Correlation between cloud condensation nuclei concentration and aerosol optical thickness in remote and polluted regions. Atmos. Chem. Phys., 9(2): 543-556. 10.5194/acp-9-543-2009Bigg, E.K. (1990) Aerosol over the Southern Ocean. Atmospheric Research, 25(6): 583-600. https://doi.org/10.1016/0169-8095(90)90039-FFossum, K.N., Ovadnevaite, J., Ceburnis, D., Preißler, J., Snider, J.R., Huang, R.-J., Zuend, A., O’Dowd, C. (2020) Sea-spray regulates sulfate cloud droplet activation over oceans. npj Climate and Atmospheric Science, 3(1): 14. 10.1038/s41612-020-0116-2Vallina, S.M., Simó, R., Gassó, S. (2006) What controls CCN seasonality in the Southern Ocean? A statistical analysis based on satellite-derived chlorophyll and CCN and model-estimated OH radical and rainfall. Global Biogeochemical Cycles, 20(1): GB1014. http://dx.doi.org/10.1029/2005GB002597Wylie, D.J., Harvey, M.J., de Mora, S.J., Boyd, I.S., Liley, J.B. (1993) Dimethylsulfide and aerosol measurements at Ross Island, Antarctica. In: G. Restelli & G. Angeletti (Eds). Dimethylsulphide, Oceans, Atmosphere and Climate. Kluwer Academic, Dordrecht: 85-94.Citation: https://doi.org/
10.5194/acp-2020-1246-RC1 -
RC2: 'Comment on acp-2020-1246', Anonymous Referee #2, 10 May 2021
This paper, by Humphries et al., analyzed data for Cloud Condensation Nuclei and Condensation Nuclei in the Southern Ocean (SO) during MARCUS and CAPRICORN campaigns covering late autumn-early spring. The results show that there are three distinct regions in the Southern Ocean near the East Antarctic coastline based on the aerosol composition and CCN abundance. Continental and anthropogenic pollution mostly affected the northern most sector. Mid-latitude CCN number concentrations and sources were influenced mostly by biogenic and sea salt aerosols in the boundary layer, and southern most sector was influenced by phytoplankton emissions. The classification of these regions is further supplemented by HYSPLIT back trajectory analysis. The results from this study clarify several questions about the sources of aerosol and CCN composition in the SO. The scop of this study is relevant to climate models for understanding the radiation and compositional biases in the SO. The paper is well constructed, methodology and measurement techniques are clearly illustrated. The content is appropriate for the context of the ACPD journal.
A few areas of the manuscript need minor revisions:
1. Figure 3. Along with the violin plots, it would be useful to plot a histogram of the number of observations for each case.
2. Figure 4. and discussion: For the southern-most sector, would it be possible to extend the back trajectory analysis to longer time period, say 15 days? Why was the back trajectory constructed only for 5 days? The authors find that aerosols and CCN in this sector are affected by aerosol species that spend significant time in the free troposphere. This explains that these fine particles have a longer lifetime and can be transported from longer distances. Therefore, extending the back trajectory analysis to longer time periods can be useful to understand their sources.
3. Figure A3 and A4. Add y-axis labels for all figures or move the y-axis label farther with larger fonts so it is common to all figures.
4. Figure A8. Is it possible to add the number of data points considered for analysis along the track? Besides , sea spray production from ship tracks can also affect the sources of aerosols due to whitecap generation. Is there a way to eliminate the amount of sea spray from ship tracks ?
5. Line 252 - Expand ERA-5. Add a few sentenced on ERA-5 data used in this study. Does the total precipitation used here refer to the model variable in the entire column? What are the spatial and temporal resolution of the data used?
6. Line 404 - Authors note that Atmospheric Compositional Front of Antarctica (ACFA) can be affected by the synoptic meteorology. What are the various synoptic meteorological processes in this region that can potentially affect ACFA and how will the sources and trajectory change? This is important for future field campaigns and interpreting data from other campaigns/seasons.
7. For the southern-most sector, would it be useful to plot a probability density / two-dimensional histogram map for chl-a Vs CCN number concentrations similar to Figure A9? This can help understand the sources of phytoplankton emissions in a better way.
Minor errors:
1. Line 1 - "The Southern Ocean region is one of the most pristine in the world" - missing word. "one of the most pristine regions in the world." Similarly, in line 29, "one of the most pristine regions of the planet".
2. Line 9" Rephrase to "where continental and anthropogenic aerosol sources coexist with the background marine aerosol population"
3. Line 37 - "impact global cloud" to impact of global cloud and carbon-cycle feedbacks.
4. Lines 42-53: Field campaign abbreviations need to be expanded.
5. Line 62: Expand on the biological processes and rephrase the sentence.
6. Line 64: "emitted from phytoplankton, into tertiary" to "emitted from phytoplankton, transforms into tertiary"
7. Line 73: "aerosol sulfur" to "sulfur aerosol"
8. Line 76 - expand abbreviation "CLAW"
9. Line 76 - "significant in CCN production" to "significant to CCN production"
10. Line 92 - A setence on the major conclusion from Alroe et al., would be useful to the context.
11. Line 263: Remove commas “for all parameters are unsurprisingly observed”
12. Line 335: Rephrase - change "-' to ','
13. Line 337: Explain air-mass fetch.
Citation: https://doi.org/10.5194/acp-2020-1246-RC2 -
AC1: 'Authors Response to Reviewers on acp-2020-1246', Ruhi Humphries, 18 Jun 2021
Author responses to reviewer comments
ACP-2020-1246
"Southern Ocean latitudinal gradients of Cloud Condensation Nuclei"
by Ruhi S. Humphries et al., Atmos. Chem. Phys. Discuss.,
https://doi.org/10.5194/acp-2020-1246-RC1, 2021
Referee #1
We would like to thank reviewer 1 for the time taken to thoughtfully review our manuscript. The manuscript has been improved in important ways in response to their comments, details of which are described below in the format used in the reviewer’s comments.
- Line 18 – we have edited this sentence to highlight the importance and need for more long-term stations.
- Line 38 and following – we thank the reviewer for highlighting this important paper. We have added it in as a reference.
- Line 195 and following – the vent emissions from indoor air and cooking, and more significantly from waste incineration, are important considerations. However, these emission sources are collocated with the exhaust emissions on both platforms, and represent a similar signal to the diesel exhaust itself, so are captured by the removal process described in the paper. We have clarified this in the first paragraph of the platform exhaust section.
- Line 315 – We have extended this paragraph to include the important results of Vallina 2006 and Fossum et al. 2020.
- Line 395 – we have altered the sentence to be more specific about the change expected in meteorological variables. We have also reviewed Bigg 1990 and Wylie et al. 1993 and while important pieces of work, we feel the main results are out of the scope of the current manuscript.
Technical/Minor revisions
- Line 212 – accepted and changed.
- Line 251 – Bandwidth selection is key to having the smoothed KDE be an accurate representation of the underlying data. Scott’s factor is a widely accepted method for calculating the optimal bandwidth of a KDE. The inclusion of this sentence is just making it explicit how the bandwidth was calculated for these data. No changes have been made.
- Line 282 – accepted and changed.
- Line 299 and 389 – we have removed the word “step-change” in line 299, and “step” in line 389, as this concept is better discussed in other parts of the paragraphs and labelling it as a step could be misconstrued.
We have added a sentence around line 295 to comment on the difference between CCN5 and CCN0.2 in that latitudinal bin.
We have added a sentence around line 300 discussing our findings against Andreae 2009. We thank the reviewer for highlighting this important result.
- Line 300 – Thank you to the reviewer for picking up this issue. We have changed the figure referencing in that sentence to avoid confusion.
- Line 400 – accepted and changed.
- Line 412 – accepted and changed.
- Figure A7 – this was an issue with the plotting script where the axes labels were not plotted correctly. This has been rectified and Figure A7 has been updated.
Referee #2
We would like to thank reviewer 2 for their considered comments. They have been an asset to the study and have led to an improved manuscript. We have responded below to each of the comments in a format consistent with the reviewer’s own submission.
Minor revisions
- Figure 3 – the violin plots already included in Figure 3 are themselves a version of a histogram. Consequently, since the information an histogram would show is already included in the Figure, we have not added a separate histogram to the manuscript.
- Figure 4 – we have limited the back trajectories to 5 days because of the uncertainty of trajectories in this region of the world. This was discussed in more detail in Humphries et al. 2016 (10.5194/acp-16-2185-2016).
- Figure A3 and A4 – these figures, along with Figures A5 and A6 have been updated according to the reviewer’s suggestions.
- Figure A8, adding number of data points along the ship-track – this information is already displayed to some degree in Figures 1 and 2. All data presented are hourly data and adding the number of data points along a track needs some form of binning, which is already performed in Figure 2. To the latter part of the reviewer’s comment, sea spray production from the ship itself is not a major concern on the RV Investigator (CAPRICORN2 campaign) because of the position of the inlet ~16.6 m above sea level and as far fore on the ship as possible. Any whitewash plume produced by the ship itself will generally fall well below the inlet. For the Aurora Australis (MARCUS campaign), while the measurement inlet is much higher (~25 m above sea level), the position of the inlet further aft on the ship’s superstructure means that the influence of these plumes on measurements is possible. However given our analysis is focussed on number concentrations, and most of the spray will be on the larger end of the aerosols sizes, their number will be relatively small, and is likely to be insignificant. In addition, the MARCUS data set only makes up a small proportion of the dataset due to the exhaust contamination, so it highly unlikely to change the conclusions of the paper.
- Line 252 – this section has been expanded to indicate the variable names and temporal and spatial resolution of the data utilised.
- Line 404
- The limiting factor in composing a two dimensional histogram plot (similar to Figure A9) of chl-a vs CCN is that unlike sea salt aerosols, which are produced at the source, secondary aerosols formed from phytoplankton emissions take a number of days to form. So except for very rare conditions, the wind conditions in this part of the world, together with the time required for nucleation and growth, mean that precursor emissions from phytoplankton occur several thousand km upwind from where they could be measured as CN or CCN. As such, composing such a plot would not be informative, and could in fact be misleading.
Minor errors
- Neither of lines 1 or 29 require alteration, as the noun of each sentence is clearly stated as “region” or “atmosphere”, respectively.
- Changed to the reviewer’s suggestion.
- Changed to the reviewer’s suggestion.
- Field campaign abbreviations have been expanded based on reviewer’s suggestion.
- Expanded based on reviewer’s suggestions.
- The reviewer’s suggestion doesn’t improve the sentence, so we have left it as is.
- This changes the topic being referred to, making it inaccurate. We have left it as is.
- The CLAW acronym is just the initials of the initial authors and expanding these doesn’t add to the interpretation of the acronym. In this case, the CLAW hypothesis is referred to throughout the literature with the acronym, without expansion. We have left this as is.
- Changed to the reviewer’s suggestion.
- This sentence has been split into two to further elaborate on the key findings of Alroe et al. related to this paper.
- Changed to the reviewer’s suggestion.
- Changed to the reviewer’s suggestion.
- This is a well-known and well-defined term in the domain, and we will leave it to the reader to look-up the definition where required.
Citation: https://doi.org/10.5194/acp-2020-1246-AC1
