General comments

This manuscript titled “African smoke particles act as cloud condensation nuclei in the wintertime tropical North Atlantic boundary layer over Barbados” by Haley M. Royer et al. is based on analyses of size-resolved particle composition (semi-quantitatively) and cloud condensation activity (quantitatively) data and dust concentration data. It concludes that African smoke is important for the atmospheric processes and cloud formation over the Caribbean. The addressed scientific question is well within the scope of ACP; the data and ideas in this manuscript are novel; and the manuscript is good structured in general.

However, the explanation of scientific methods including details in experimental settings and data analyses and uncertainties of the size-resolved chemical composition is unconvincing. In addition, the writing and interpretation especially in the Results and Discussion section are rough and need to be revised. The authors thus need to make careful revisions and corrections to improve the overall quality of the paper for publication in the journal. I would recommend the editor reconsider the manuscript only after a major revision by the authors.                                                

Specific comments

1. Lines 90-102: whereas the possible compositional changes of smoke particles have been well explained the explanation of morphological changes is in lack. It’s better to add the latter, too.
2. Lines 102-104: This sentence should be rewritten.
3. Line 185: since “EDX is considered a semiquantitative method” the uncertainty of this method in determining the size-resolved chemical composition should be elaborated because it directly influences the conclusions.
4. Line 188: “organic material, sea spray, dust, and anthropogenic emissions” are not parallel categories. Please revise.
5. Lines 192-194: “Samples collected on Si … generate a background signal as well.” Does this mean except for the “carbon-coated copper grids” you also used silicon filters to collect particles for CCSEM/EDX analysis? This information should be added.
6. Lines 198-201: “The algorithm operates by … and existing literature.” The details in the K-means clustering analysis and the assignment of particle types should be explained at least in the supplement. How the number concentration-size distributions in Figs. 5 and S4 were determined should also be explained.
7. In the section “Size-Resolved CCN Measurements and Data Analysis”, the details in the experimental settings including flow rates of the CCNC, CPC, DMA, and SMPS should be explained.
8. Lines 243-244: “Calculation of activation … in Pöhlker et al., 2016.” These should be explained at least in the supplement.
9. Line 255: Should “Sampling Conditions” be replaced by “Air Mass Characteristics”?
10. Lines 260-267: please restructure this part. For example, only from Fig. 1, one cannot see the result described in lines 261-262. In addition, it is better to describe the figure before it is referenced.
11. Lines 269-271: “suggesting that … onto larger transported continental aerosols (Tomlin et al., 2021).” This part needs to be modified. For example, “suggesting that transported particles overwhelmed the background marine particle loading so that the small Aitken mode particles possibly coagulated onto the larger transported continental particles (Tomlin et al., 2021).”
12. Lines 291-346: it’s better that the morphology of aged sea spray, mineral dust, and sulfate particles are also explained (and interpreted from the viewpoint of transport histories (e.g., history of atmospheric moisture conditions) if applicable).
13. Lines 299-305: the explanation on aged sea spray is not enough. Based on the explanation, one cannot see if the related particle and spectra presented in Fig. 3 is of aged sea spray particle or not.
14. Lines 315-317: the existence of N or ammonium cannot be seen from the related particle and spectra presented in Fig. 3.
15. Lines 325-327: how about the morphologies of the smoke particles in this study, which type is dominant?
16. Lines 330 and 334: please correct the elements of dust components.
17. Lines 341-342: “This scarcity of … respectively.” This sentence is incomplete.
18. Lines 362-364: “These finds … for smoke (Figure S2 and S3).” First, the origin of the data in Figs. S2 and S3 need to be explained. Next, can the CO variation during the observation period of this study indicate the arrival of smoke particles during the three CAT events?
19. Lines 378-382: please rewrite this part to make it concise.
20. Lines 391-393: “A large decrease … observed in Figure 2.” This part needs more explanations. Do the authors know any existing study that reported the condensing of marine biogenic sulfur precursors onto transported particles? Do the authors mean “The large decrease in sulfate number fraction during CAT events might be caused by the condensation of marine biogenic sulfur precursors and/or Aitken mode sulfate-containing particles onto the large, transported particles as indicated in Fig. 2.”?
21. Lines 399-401: please explain “Bin sizes for each decade” and “bin size”; please rewrite this sentence.
22. Line 410: why these small particles are less hygroscopic should be tentatively explained.
23. Lines 414-423: the difference between marine conditions and smoky conditions should be quantitatively explained and the data should be tabled for all S conditions.
24. Lines 418-420: the first half and the second half of the sentence are repeating each other.
25. Lines 428-432: “However, one finding of note … African aerosols.” Please rewrite this part to make it concise.
26. Line 431: “such as dust that activate as CCN”. Dust might haven’t acted as CCN but it still could have influenced the activation diameter and thus the overall estimated kappa. I suggest deleting “that activate as CCN”.
27. Lines 441-443: the existence of organics may have lowered the observed kappa. Please rethink this conclusion.
28. Lines 444-447: please explain the box plots in detail and the meaning of the whiskers of d₅₀.
29. Lines 458-459: “In Fig 7b, there is a clear and direct relationship” should be rewritten. For example, “Figure 7b shows a positive correlation”. In addition, this positive correlation should be further elaborated with the data for example in Table 1.
30. Line 460: “may act as CCN” should be “may have acted as CCN”.
31. Lines 464-466: “The large number of smoke particles … on sulfate or organic particles.” Do you want to say that the number of sulfate and organic particles were not enough for the measured CCN counts?
32. Lines 469-472: what are the possible surfactants? Could you explain?
33. Lines 472-479: First, can any evidence be given from the single particle composition data that aging of particles occurred. Second, what you want to conclude from this part is not clear.
34. Lines 485-486: “It is likely that … of factors.” seems redundant. It’s better to delete it.
35. Lines 450-488: the writing and organization of this paragraph are quite rough and should be revised.
36. Table 1 should be modified according to https://www.atmospheric-chemistry-and-physics.net/submission.html#figurestables
37. Line 508: the activation diameter should be explained with the measured S condition.
38. Line 513: “Overall, we find that smoke has a larger effect on CCN number concentrations than dust.” Does this conclusion apply for all time or for the studied period or season?
39. Line 522: what’s the difference between AOD and AOT? And where are the data of Figs. S2 and S3 from?

Technical corrections

1. Line 73: Should “though” be replaced by “however” or other similar phrases?
2. Line 143: “composition” should be “compositions”.
3. Line 182: “affects” should be “affect”.
4. Line 218: “Rose et al., 2008” should be “Rose et al. (2008)”. Please modify similar expressions in other places.
5. Line 227: “0.71” should be “0.71 %”.
6. Lines 251-252: “single particle data (e.g., CCSEM-EDX), and air mass history (e.g., NOAA’s HYSPLIT model)”, the “e.g.,” in the parentheses should be deleted.
7. The color shadings in Fig. 2 should have been explained in the caption.
8. Line 350: “represent” should be “present”?
9. Line 362: “observe” should be “observed”.
10. Lines 369-370: “wildfire smoke appears to overwhelm the number fraction of the submicron aerosol loading” should be “wildfire smoke particles appeared to dominate the number fraction of submicron aerosol.” Please also correct the wrong use of overwhelm in other places throughout the manuscript, e.g., line 510.
11. Line 388: “Figure 4” should be “Figure 5”.
12. Line 452: “N Atlantic MBL” should be “North Atlantic MBL”.
13. Line 455: “Fig 7a suggests” should be “Figure 7 suggests”. Please refer to https://www.atmospheric-chemistry-and-physics.net/submission.html#figurestables
14. Line 476: “to oxidants” should be “to oxidize”.

This work reports size-resolved single particle composition and cloud condensation activity results during the EURECA and ATOMIC campaigns from January to February 2020 over the tropical North Atlantic. They concluded that aerosol hygroscopicity and CCN number concentrations during wintertime long-range transport events over the Caribbean are affected by African smoke more than dust. The main content of this manuscript is well within the scope of ACP. The manuscript is well-prepared in general. However, as the other referee commented, the writing and interpretation of your results, especially the part to support of your major conclusion, need to be carefully revised. I would also recommend the publish of this manuscript after the following issues are fully resolved.

General comments:

A major part of your results is size-resolved chemical composition, which you used to support your main conclusion. However, from your methods, I did not see a clear explanation how you obtained this “size-resolved” information and it should be an really important part for your methodology. As you stated “EDX is a semiquantitative method”, then how did you ensure the accuracy of the results in Fig. 4, Fig.5 and Fig. 7, and eventually drove your conclusions?   

According to Fig.7b, you stated the CCN counts correlate quite well with smoke number fraction. Could you also plot similar figure for dust number fraction with CCN counts?

Continued with my comment above, you only measured the chemical composition of particles at diameter < 0.7μm. You also agreed that there might be more dust particles in the supermicron range. These particles may also be good CCN as they are large enough. However, you did not measure those. From this perspective, I might not agree that smoke has a larger effect on CCN number concentration than dust. You should consider how to re-interpret your results.

On page 24, you gave several reasons that smoke particles may act as CCN. However, I did not really agree with those reasons you provided starting from line 466 til the end of this paragraph. First, you did not have any direct evidences for these explanations. Second but more importantly, you just mentioned in line 405 that smoke particles lower submicron aerosol hygroscopicity. Then these particles can act as CCN would not be due to the elevated hygroscopicity or κ, which arises from the presence of water-soluble compounds, for instance WOSC, aged organics or other salts.

Specific comments:

Line 161-162: BC particles did not evaporate at 500 °C, either. How did you exclude them?

Line 185-203: Please provide detailed information how you obtained size-resolved chemical composition for particles.

Line 202: Similar to my general comments, since you studied particles at size below 0.7μm, where dust might not be as abundant as smoke, then how did you drive the conclusion about the dust effect on CCN activity?

Line 246-248: How did you prove current statement? From which figure? This sentence fits better into conclusion part, but not the starting of results. Again, how did you identify smoke particles? It came up without any description, though you explained it later afterwards. Please consider rephrase your results structure.

Line 266: Were the Aitken size mode particles from the marine background? Why?

Line 315 and 317: strong of what? Please rephrase it.

Line 325 to 327: Then what is the morphology of smoke particles in your study? You should provide more information of your current results.

Line 354: How did you obtain the number fractions of each kind of particles? Please add this part into your methodology.

Line 366 to 370: Please consider rewrite the sentence, it is confusing.

Line 381: From Fig. 5, I cannot directly see those particles are the largest ones. Maybe you can plot a figure for the normalized mean diameter of each particle type, which is more obvious.

Line 382-384: I cannot understand what do you mean “clean” here, please clarify. The number concentration or mass concentration of particles were lower for clean marine conditions?

Line 408: Sometimes you used “dusty conditions”, sometimes you used “continental influenced aerosols” or sometimes referred to “CAT Event”. At line 415, you used “smokey conditions”. Could you please be consistent?

Line 427 to 428: Did they obtain those values at the same S? Otherwise, you cannot directly compare them.

Line 454: Could you compare the results at those two CAT Event or for those three Clean Marine Period? For instance, for those two CAT events, the CCN concentration was different with different hygroscopicity, even though they are both influenced by the continental transport. Could you please give the potential reasons and make proper discussions.