Dear author, please perform the corrections suggested by one of the reviewers of your paper and let me know how these corrections have been incorporated in the further revised version of your manuscript.
See the list of comments at the end of this message.
I am looking forward receiving the new revision of your manuscript.
Maria Kanakidou
--------comments by the reviewer ------------
I am in general satisfied with the rebuttal of Hauglustaine et al. to my review. I have re-read the manuscript and supplement, and list some minor comments that I noticed in the revised version. It would be good if these last comments could be taken into account before accepting the paper.

l. 75: Explain what Bauer et al. found- as relevant for this paper.
l. 83 Give numbers between brackets.
l. 106 irreversible removal ?
l. 149-151 Perhaps it is opportune to explain that the small IH mixing time points to relatively vigorous (local) vertical mixing. Perhaps the authors can also refer to published radon simulations?
l. 267 Do you refer to a specific ISORROPIA version? Here you mention ‘fairly good’, in the appendix it is ‘exellent’. I suggest to summarize this as follows:
. an excellent agreement between both model results has been over a range of key parameters currently encountered in the global model version, with some larger deviations (>xx %) at temperatures larger than 295 K.
l. 271 this is somewhat slang. Numerically, first photochemistry ….
l. 446 It is interesting to see the diurnal variations in the appendix. However, the authors should try to include a somewhat more extensive description of what day-night ratios are found in these regions. I understand you don’t want to run into an extensive comparison, but only results from one paper is somewhat meager.
501 As illustrated by Xu et Penner=>do you want to say: to enable comparison with Xu and Penner???
502 I don’t understand these negative ratios. Is this something numeric? As it is written in the text the numbers can not be negative. I suggest to avoid confusing language- and figure by modifying the color scale (grey value above a small threshold value, white colors smaller than this threshold). It is probably not important, but should be correct.
525 (or 8 &) of all nitric acid formation.
532 Also mention that loss onliquid (and ice) clouds were not considered. To some extent cloud occurrence puts some lower limit to the N2O5 removal- if there is a cloud the reaction is almost certainly fast. The removal rate is then mostly dependent on the statistical probability that air masses are mixed with clouds, timescale in the order of a few days.

559 If the authors agree I propose upgrading of this statement (towards conclusions/abstract) as the number has not often been recalculated since Dentener/Crutzen and Lee et al paper.
D.S. Lee, I. Kohler, E. Grobler, F. Rohrer, R. Sausen, L.Gallardo-Klenner, J.G.J. Olivier, F.J. Dentener, and A.F. Bouwman, Estimations of global Nox emissions and their uncertainties, Atmos. Environ., 31, 1735-1749, 1997.

606 The description in the supplement of deposition is now somewhat ‘widowed’, and could refer back to this section (or include some of the main text)
722 and further: please check units microgram/cm3=>microgram m-3???
757: negative ratios: avoid to introduce this concept.
Supplementary material:
l. 46 essentially=>hardly anywhere?
l. 50 regions concerned=>regions with fine nitrate formation
l. 53 dissociate is not the right term. Evaporate?
l. 53 A somewhat more extensive linkage to available studies on day night ratios is needed.
98 there is virtually no choice=>it is not possible to rely on
110 In Figure S5 etc: does ‘world’ include or not Europe/North America.
113-116: more is explained in main text. Description here is almost absent.
Figures S4-S6: The lettertype is hardly readable, too small in the pdf.

Dear authors:
Thank you for submitting this nice work to ACP and for the careful revision of your manuscript that can be now published in ACP,
with kind regards,
Maria Kanakidou