Overall, the paper clearly improved, but there are still many typos and smaller errors left. In particular the new parts of the manuscrip have numerous spelling, punctation, and other small errors: e.g. P1L24, P1L25, P2L3 (terestrial + buget), P2L9, P2L28, P2L33, P3L15, P5L4 ('an data'), P5L6, P7L22, P8L19, P10L25, P11L4

The authors should carefully work through the manuscript and improve the points mentioned above. In addition, they should also address the specific comments detailed below.

Specific comments:

P1L24-25: "Comparing the AOD properties" doesn't make much sense here.

P3L6: "aerosol size parameter" should be replaced by "size-related aerosol parameters".

P3L7: "air mass Caribbean-African connection climatology" should be improved.

P3L17: "zone" should be replaced by "region" or "area" (also in Fig 1, 8, etc) as suggested by one of the reviewers in the initial reviews

P3L23: Please clarify here (and in Fig. 5+9) that the VPSD means dV(r)/dln(r).

P3L24: "Real Refractive Index" should be replaced by "real part of the refractive index", same for the imaginary part.

P4L4: "no loss of accuracy" is probably too strong and according to Tab. 2 of Mateos et al. 2014 the statement seems to be wrong as there is some loss of accuracy.

P4L5-6: here the abbreviations like REFR etc. could be used

P4L8: "in the ... sites" seems a bit unusual to me. I would suggest to write "at the ... sites" instead. This occurs many times in the paper.

P4L12: "for Barbados" does not fit well to step 1 and 2 because the description of the methodology is more general. So you should describe step 1 and 2 using the example of Barbabos.

P5L14: it should be mentioned that case c) also includes NoD_AF+D_CAR even if it does not occur often. Or just write "c) NoD: No-dust conditions at the Caribbean sites".

P5L15: "showing connection" -> "with connections"

P5L18: "however" is repeated.

P5L27-28: This sentence is not unclear. Please clarify. Maybe the sentence also could be removed.

P6L15: "large" -> "largely"

P6L23: "As a first step we show" -> "In Fig. 3 we show"

P7L15: it is probably sufficient to say "at about 2 µm" given the inversion uncertainties and the resolution of the size grid. This occurs also at other places in the paper where it also should be corrected.

P7L24: "achieved at larger radius (about 3.85µm)" -> "shifted to larger radii (> 3µm)"

P8L5: The title of section 5.1 is not well choosen. I suggest something like "Overview over air mass connections".

P8L18: You should add here that you mean level 2 of AOD and AE data.

P9L9: Maybe "misidentification" is not the best word because they were identified correctly as AOD < 0.2. I suggest "probably there are days with presence of dust with AOD ..."

P9L10-11: I think this sentence could be removed.

P9L12: "so as to lower values" could be removed.

P9L14: The title of section 5.3 is not optimal. If "Scatterplot" is left in the title of Section 5.2, it could be rephrased asr 5.3 "Scatterplot of connected AOD data"

P9L25: "between" -> "of the" ; "leaving" -> "between"

P9L31: "etc" is not good, "for example" might be better.

P10L3: "the VPSD" -> "the average VPSD"

P10L9: "is a loss about 35%" -> "is a loss of about 35%"

P10L23: I think "(defined by aspect ratios)" is not necessary here.

P10L31: "during SALTRACE" -> "during the SALTRACE"

P11L21: " the reported " -> "those reported"

P11L28-29: "quite similar from event to event": Where is this shown is the paper?

Fig 3 legend: "AODCAR and AECAR": CAR as subscript; also not correct at other places in the paper

Fig 4 lenged: "of:" -> "of"

Fig 6: "Monthly/Annual mean Links" might be improved to "Average number of links per month/year"

Fig 6: The labels on the x-axis should be homogenized to the labels in Fig. 2+3

Fig 7 legend: ", , and grey triangles are non-dusty days (“NoD”) in both areas": This doesn't make sense because there are grey triangles in the "dusty region" of subplot b

Fig 8 legend: "bar" could be removed; "DAF+DCAR" subscript missing; you should add that the plot is for days where a link was found.

Fig 12 legend: It should be mentioned that these are average values.

In general the figure legends should be homogenized (e.g. write everywhere "DAF + DCAR case" where applicable , e.g. in Fig. 11 and 12

1) In the abstract, please change "We observe that AOD at 440nm decreases about 0.16 or -30% after transit." to "We observe that the AOD at 440nm decreases by about 0.16 or 30% during transport."

2) In the legend of Fig. 6 the first "total" should be removed.

3) In Fig. 8 "r^2=0.04" is written. It is not explained what this means. Please clarify. (In Fig. 4 "R" is written which probably is the correlation coefficient.)