Dear author, co-authors, I have read the reviews of your ms as well as your response to the comments provided by both reviewers and your revised version of the ms. One of the reviewers was apparently appreciating the contents of your paper and only had some minor comment which you addressed well in your response and revision of the ms. The other reviewer had some major criticism on especially the explanation/justification of issues on the emissions and role of vertical mixing in explaining some of the model biases in surface ozone in the SE USA. In response to these comments you bring up an interesting reasoning that to some extent might indeed explain why models such as GEOS-CHEM overestimates surface ozone in this region of the US. However, it also raises a number of additional questions that should be addressed; You are indicating that one of the main reasons for the overestimation of surface (or boundary layer) ozone in GEOS-CHEM is the apparent too dry input meteorology. This would according to your explanation result in a too strong chemical production of ozone as well as too strong boundary layer mixing. How would you diagnose that? You indicate that the boundary layer mixing has been modified to result in the simulation of a shallower BL depth in agreement with the observations but assessing boundary layer mixing requires to also assess vertical profiles in potential temperature, moisture and concentrations of tracers such as ozone. The latter explanation on the too dry conditions and impact on BL mixing then introduces some controversy since with a too strong boundary layer mixing you would detrain/export some of the ozone precursors much more efficiently out of the boundary layer (as well as the excessive ozone if this would be indeed higher in the BL compared to free tropospheric ozone levels). But also now that you bring in this explanation for the overestimation of surface ozone, it triggers a major new question: how does the too dry meteorology in GEOS-CHEM affect atmosphere-biosphere exchange processes. The too dry (and warm?) conditions might strongly enhance some of the BVOC and soil-biogenic N emissions, although you indicate that the latter term is probably of minor importance compared to the anthropogenic emissions. It is indicated in the paper that some of the BVOC emissions have been scaled down, a correction needed to deal with the bias in the input meteorology? The other issue that requires than further analysis and discussion; how is the dry deposition is affected by these too dry meteorological conditions in the model? You indicate that GEOS-CHEM uses an implementation of the Wesely scheme and where you have indicated that there have been some modifications to properly simulated the observed dry deposition for one specific site in the US. How much is the overall dry deposition for the whole region changed by these modification and how much does this affect surface ozone levels compared to some of the other scalings, e.g., in NOx emissions and boundary layer depth? It is interesting to note that you would actually expect that, due to the too dry conditions you would underestimate dry deposition when you would consider the potential impact of soil moisture and other hydrological drivers of vegetation uptake such as vapor pressure deficit on dry deposition but I assume this is not included in GEOS-CHEM.
One of your main conclusions is that the EPA NOx emission inventory seems to give a too large source of NOx for the US. Given the potential implications of this finding I appreciate the comments by the second reviewer to secure a further support of your case that it is indeed the emissions that are overestimated and not a selection of processes that are not properly represented in your modelling system. I do though also see that this could easily result in a couple of further follow-up studies, as you also indicate in the conclusions. Including in your discussions some more explicit description on what would be needed to further corroborate your findings would make a further valuable modification of this ms. My specific feedback is that I deem being essential is to conduct some detailed analysis for some specific locations (with field observations with chemistry and micro-BL meteorology observations) to assess the diurnal cycle in dry deposition (not only of ozone but also precursors and products), boundary layer evolution (not only BL depth but also vertical profiles) and analysis of the diurnal cycles in the ozone process tendencies.
I anyhow intend to invite the second reviewer to provide an evaluation of the revised version of the ms for making my final decision but before doing so, want to raise these points that you can potentially include in a further updated revision/response.

Regards, Laurens Ganzeveld

Dear author, co-authors, I have received now a second review of your ms submitted for publication in ACP. As you will read in this second review, provided by one of the reviewers of the first version of the ms, there appear to be still a number of issues that I would like you to address in a further revision and reply,

Regards, Laurens Ganzeveld

Dear author, co-authors,

After some delay (was out of office for two weeks), I can provide you my final decision on the paper. Having read the response to the reviewers comments and also considering the whole iteration process I now accept the paper for publication in ACP after the last minor issues have been tackled.

In your response you indicate that you have included now as an extra analysis one that assumes zero soil NOx emissions. This is clearly a drastic sensitivity analyis where which you justify referring to the large uncertainty in soil NOx emissions. I suggest to change some of the statements about this extra feature of your analysis:

Discussion: "Our results indicate that NEI NOx emissions from mobile and industrial sources must be reduced by 30-60%, the range reflecting uncertainties in soil NOx emissions"

Change to: "Our results indicate that NEI NOx emissions from mobile and industrial sources must be reduced by 30-60%, dependent on the assumptions on the contribution by soil NOx emissions"

Page 13, line 20: “Soil and fertilizer NOx emissions (18% of total NOx emissions in the Southeast) and open fires (2%) would be insufficient to correct this bias.”

Change to: “Assuming strongly reduced soil and fertilizer NOx emissions (18% of total NOx emissions in the Southeast) and open fires (2%) also considering the large uncertainty in these emissions would be insufficient to correct this bias.

Line 25: “We can achieve the required 40% decrease in total NOx emissions by reducing NEI emissions from mobile and industrial sources (all sources except power plants) by 60%, or alternatively by reducing these sources by 30% and zeroing out soil and fertilizer NOx emissions. Here we choose to do the former. There is enough spatial overlap between anthropogenic and soil emissions that we cannot readily arbitrate between these two scenarios.”

Change to: “We can achieve the required 40% decrease in total NOx emissions by reducing NEI emissions from mobile and industrial sources (all sources except power plants) by 60%, or alternatively by reducing these sources by 30% and zeroing out soil and fertilizer NOx emissions. Since it is apparent that there is some minimum contribution by soil NOx emissions we assessed the impact of the approach reducing the NEI emissions by 60%. The spatial overlap between anthropogenic and soil NOx emissions is such that we cannot readily arbitrate between these two scenarios”

Conclusions: “, we find that emissions from other industrial sources and mobile sources must be 30-60% lower than NEI values, where the range reflects uncertainty in the contribution from soil NOx emissions.”

Change to: “, we find that emissions from other industrial sources and mobile sources must be 30-60% lower than NEI values, dependent on the assumptions on the contribution by soil NOx emissions.”

Finally, I see that you have removed the statements about the “excessively dry conditions in GEOS-CHEM”. This was also based on the comments raised by the reviewer which was hinting towards a more detailed analysis then of some of the BL meteorological properties to further assess this. However, reading again the revised paper, I would actually suggest you to keep this indication about a potential explanation of some the issues on the simulations of too high ozone in the modelling system since it nicely indicates what should be focus of future research. I would suggest to clearly state that (something like) “It is beyond the scope of the presented study to further analyze this in more detail. However, also recognizing the role of the hydrological cycle in surface and boundary layer exchange processes such as soil NOx emissions, dry and wet deposition and turbulent transport, this feature should be focus of follow-up studies regarding assessments of air quality over the US using GEOS-CHEM”

Regards, Laurens Ganzeveld

Dear author, co-authors, after quite some iterations I accept your ms for publication in ACP. I hope it will trigger further studies on the air quality over the US and where it will be especially interesting to see how some of these features of the representation of the hydrological cycle and its impact on atmospheric chemistry processes will be further subject of such follow-up activities,

Congratulations and regards, Laurens Ganzeveld