Dear authors, co-authors,

I have read in detail once more again the revised ms as well as the reviews, including the one public comment and your response to these two reviews and comment. Overall, it seems that you have tackled well the various raised comments and that the revisions you have included in the manuscript properly addresses some of the main criticism. My own criticism is that there are a number of occasions where the actual models response cannot be well explained leaving you to hypothesize what might be the explaining mechanisms behind that response, e.g., changes in chemistry over the EU, a stronger response in transpiration compared to the Lombardozzi et al. (2015) study, the reduced N-resource limitation. This is also further confirmed in your response to, especially some of the comments raised by reviewer #2, e.g., the link between CUO and surface ozone concentrations and the explanation for the different responses in isoprene in the various regions. It calls according to me for a more detailed analysis for specific sites, e.g., with 1-D model approaches of the similar system so that you can really nail down the various explanations for all these mechanisms. That would, however, be enough material for another paper and appreciate this paper providing a strong motivation to indeed further pursue such further in-depth process understanding studies also for those regions where these interactions and resulting feedbacks mechanisms seem to be most relevant.
In this process of reading over the comments, your response and the resulting revisions, I still came across a couple of editors comments that I would like you to address for my final decision.

You could consider to include the reference by Super et al., 2015JG002996, Cumulative ozone effect on canopy stomatal resistance and the impact on boundary layer dynamics and CO2 assimilation at the diurnal scale: A case study for grassland in the Netherlands, J. Geophys. Res. Biogeosci., http://dx.doi.org/10.1002/2015JG002996, after the statement on the impact of entrainment on ozone, lines 91, 92

Line 176: “multiplied with..”

Line 184: “Therefore, online ozone-vegetation coupling and feedback are included”, I would propose here to state that the “previously discussed feedbacks are mostly included” also since you might not cover yet all the potential feedbacks (like ozone deposition impact on NOx exchange).

Line 220; this finding that the ozone impact in the simulations with an interactive calculation of ozone concentrations and using prescribed ozone is an interesting one since it already suggests that the overall impact of the coupling is not that large, resulting in substantially differences in ozone, or is it more a coincidence dependent on what prescribed ozone climatology you used?

Line 230-232; Discussing your results by comparison with other global scale climate-chemistry studies; were you aware of the Ganzeveld et al. 2010 study on the impact of land cover and land use changes on O3 and atmospheric-chemistry climate interactions? I am mentioning this since in that study we showed that the impact of future land cover and land use changes on ozone was small also due to the role of compensating effects by inclusions of some of the same interactions as you consider in this study, e.g., changes in LAI (line 258), micro- and boundary layer meteorology and biogenic emissions, except of the ozone uptake impact.

Line 280: “biogenic isoprene (or VOC?) emissions”; I think that you should stress that you consider only changes in biogenic emissions of isoprene and did not consider changes in biogenic N-emissions. By the way, so far you mainly referred to isoprene emissions but do changes in biogenic emissions also include changes in terpene and other BVOCs emissions? This is not completely clear. You should possibly stress this.

Line 287: “which provided the sensitivity of simulated ozone to perturbed dry deposition only without the complication of stomatal coupling with hydrometeorology”. This modification doesn’t read well. I would propose to rephrase this to: “which provides an approximated sensitivity of simulated ozone to perturbed dry deposition velocity only to separate this impact from that due to changes in hydrometeorology associated with changes in stomatal conductance, e.g., changes in mixing layer depth”. This also links this statement better to the follow-up discussion/statement on mixed layer depth changes. I also think you should carefully refer to the term dry deposition being or the dry deposition velocity or dry deposition flux. Since you are referring in this part of the analysis to figure 5a, it should all read “dry deposition velocity”.

Line 291-292: “…feedback whereas over western Europe, where the lower chemical loss rate following reduced transpired water might have further enhanced the positive feedback”; see my suggestion for sentence change but also want to remark that such a statement about what might have caused a stronger response does not give a strong impression about the level of understanding of the actual explaining mechanisms. Are not there not additional diagnostics to indeed confirm that it is change in the chemical loss rate that explains the simulated response?

Line 314; here I read for the first time a reference to the role of soil moisture changes whereas I think this is an extremely essential component in this coupling mechanism. Has there indeed been hardly any change in soil moisture, has it not been focus of your study or was it already included in more detail in the other preceding studies? I am inquiring, also since I am informed that for example the soil moisture signal might be an essential component of the Mediterranean pollution response (the ecosystems being much less sensitive to ozone when stomatal closure due to reduced soil moisture is considered, see studies by Emberson et al. on the DOSE modelling and EMEP).

Regarding some of the revised figures, also the ones you included in the supplement; there is one striking one, Figure S3 which shows the relative changes in the O3 dry deposition velocity, indicating differences of -20% over Greenland. I am aware that there it is reflecting a relative difference for a very small absolute difference in Vd, which is mainly reflecting the snow-ice uptake rate and the turbulent and diffusive transport to the surface. You would guess that the snow-ice uptake would not change in your model (which is the case in most models) and that the change would reflect a small change in Vd in turbulent and diffusive transport. This is possibly also reflected in quite large differences in surface temperatures and heat fluxes expressed in Figure S4. Anyhow, so showing these relative differences including this “surprising” signal in such areas calls or for a better explanation of the mechanism, or you might prefer to only show the relative differences for those regions where this is some significant vegetation cover. This would serve to avoid getting too much distracted from the main point you want to make and requiring a more detailed explanation of such additional, complicated system responses.

I also checked the revised Figures 6 which were changed to tackle one of the comments by one of the reviewers about statistical significance resulting in a recommendation of major revision. The dots should indicate the regions where the changes are significant but have to say that these came out poorly and were only visible by a very careful check of the figures. I recommend that you still try to find a more optimal way to indeed indicate the significance levels, e.g., only showing the contours for those locations where the signal is indeed significant.

Regarding this specific comment by reviewer #1:
p8 l275-277: As mentioned above in connection with figure 6, I don’t quite understand the line of argument here. Why can the ozone impact on its own deposition not be diagnosed directly from the model simulations? Please explain further.
Your response: See our response to point (3) above.

I was also wondering about this point. I have already provided some additional comments on how you tackled this comment #3 and where I understand why these diagnostics might be complicated. However, I don’t think it is computationally complicated. It points to me that you have not available in your output diagnostics the separate terms that all at the end determine the actual change in Vd which are Ra + Rb + Rsurf.. If you would have those terms written out you could directly calculate the actual resulting changes in Vd and the diagnose the contributions by the micro-met terms relative to the surface uptake term. Somehow, the currently provided explanation does not seem to be very satisfactory and would still like to invite you to see if this can be more optimally tackled.

I hope that this last round of editors comment will help in arriving at a final version of the ms that optimally describes the main outcome of your study and that will further trigger research with a focus on this theme of ozone deposition impacts, interactions and feedback mechanisms.

Dear author, co-authors, have checked in detail your response to the last round of comments as well as the revision and do accept now this paper for publication in ACP. By the way, it seems that the topic on O3 deposition impacts and beyond is currently receiving more and more attention with another paper on this topic being submitted to ACPD,

Regards, Laurens Ganzeveld