Dear authors,

Thank you for your revised manuscript. The reviewers of your initial submission recommended either technical corrections or minor revisions. From my reading of this revision I feel you have addressed these comments; however, I have a number of small language issues from the earlier portion of the study I’d like to raise. These are detailed below:

Line 10: the title of the paper says 2002-2018 but this line says 2011-2017, which is a bit confusing. I understand that Ny-Alesund has (regular) data for 2002-2018 and Barentsburg for 2011-2017. So I would reword to better state in the Abstract that the analysis covers 2002-2018 but the period with two sites is only 2011-2017.

Line 12: “It was found that generally the fine mode aerosol optical depth dominates and also shows a larger interannual as inner annual variation. The fine mode optical depth is in fact larger in spring during the Arctic Haze period.” These sentences do not make sense: in the first one, “interannual” is repeated twice, and in the second one, from later Figures it looks like fine mode optical depth is larger than coarse year-round (not just in spring). I wonder if the first sentence is meant to say that fine mode has larger interannual and “seasonal” variability than coarse mode? And the second that fine mode is “largest” (not “larger”) during the spring?

Line 15: ”Overall the aerosol optical depth seems to decrease, although this is not statistically significant.” I think it is useful to put the estimated trend and its uncertainty in the Abstract. It should also be included in the paper (I saw it mentioned but didn’t see numbers written down somewhere). Seeing the estimated value and uncertainty are both important. A trend can be insignificant because it is very small, or insignificant (but potentially important) because the uncertainty is high. I do not think it is possible for the reader to tell from the information provided within the paper.

Lines 107-108: is the correct version of HITRAN specified here? The latest is HITRAN 2016. Also, the version of the MT_CKD continuum model used should be specified. This will have an effect on the midvisible and swIR bands.

Line 136: I am not sure that “juxtaposed” is the right word here. Are you saying just that you compared the data sets, or something else? This is not clear to me.

Line 144 (and maybe elsewhere, e.g. 209): I would refer to 0.87 microns as near-IR, rather than IR.
Lines 168-169: Unless I am misunderstanding here, beta is defined as equal to the AOD at 1 micron (not just “close to” it).

Lines 171-172: I am not sure that the Eck and Schuster references fully support this. They point out that a quadratic formulation (in log-log space) is more realistic than a simple power law when going into the UV (e.g. 0.34 microns as done here) when the fine mode is optically dominant (as is the case here). It is not just at the longer wavelengths as described in your next paragraph. Maybe an extra sentence can be added about that.

Please let me know if you have any questions- I don’t think another round of review will be needed after the above comments are addressed.

Best wishes

Andrew

Dear authors,

Many thanks for submitting this further revision with the minor corrections and clarifications I had noted. I have read through and am pleased to accept this version of your study for publication in AMT.

Best wishes,

Andrew