The manuscript discusses aircraft observations from three campaigns over Europe. One campaign was conducted in May/June 2020, when lockdown induced emission changes provided a unique opportunity to study changes in atmospheric chemistry. The study shows that emission changes of NOx had a profound impact on tropospheric ozone production regimes. The paper is well-written, but would benefit from further detail regarding the methodology and simplifications that represent the basis of the analysis. In this context, the fact that no accompanying VOC measurements were conducted presents a shortcoming that should be addressed in more detail.

Detailed comments:

Introduction:

The introduction would benefit from a more robust literature review on early papers that have unravelled the relationship between NOx and O3 chemistry.

Eq. 1: Why can RO2 cross reactions be ignored here? A rational should be given why these terms are omitted for the lower continental atmosphere (e.g. <4km). CH3O2 is exclusively produced from methane oxidation, however even in the remote atmosphere, methane and VOC OH-reactivity are comparable (see: Mao et al., ACP, 2009. doi: 10.5194/acp-9-163-2009). The presented simplification might work for the remote marine atmosphere, but I doubt that it is applicable to polluted continental areas in a quantitative sense as analysed here.

Line 73: The authors mention emission reduction studies were only performed at the surface, but none of the cited papers actually investigated emission reductions. The cited studies investigated changes in ambient concentrations, which are typically subject to chemistry and meteorological / climatological differences. It is somewhat unclear what the authors try to say here. Aren’t pollutants (with a few exceptions) primarily released at the surface – or do the authors rather want to refer to the impact of emissions on atmospheric chemistry? If the authors specifically mean that emissions released above the surface (e.g. from air traffic) are of importance, I would suggest to reword this paragraph and be more specific about this.

Line 100: ok here methyl peroxy radicals are simply based on methane, but earlier (line 47), methanol was also mentioned as an important precursor for a study site in Finland. In fact there could be many more precursor VOCs for methyl peroxy radicals  in the upper atmosphere (e.g. the photolysis of carbonyls, or subsequent RO2 x HO2 and RO2 x RO2 reactions of most carbonyls)

Using a campaign in Africa as a reference seems a stretch here. What about the role of biogenic VOCs and oxidation products? For example Crutzen et al., (Atmos. Environ., 2000: doi: 10.1016/S1352-2310(99)00482-3) found significant amounts of BVOCs and their oxidation products in the tropics up to 10 km.

What is the bias of neglecting other RO2 sources (e.g. changes in anthropogenic VOCs and BVOCs) over Europe? Also, May / June represent seasons where biogenic emissions in Europe should play an increasing role.

Line 145: So the campaign was conducted in May/June, when most lockdown related restrictions were already easing in Europe – would the analysis presented here then be more of a reflection of the post-lockdown regime, with some restrictions (e.g. travel restrictions) still in place, others not? For example traffic volumes across Europe and elsewhere (e.g. China) largely recovered by June 2020.

Line 144 ff: No VOCs were measured during these campaigns, which presents a major uncertainty. At the minimum the authors should state something about anthropogenic VOC emission changes and estimate the potential change in VOC reactivity prior and after the lockdown  relative to NOx. Observations of lockdown induced reductions of anthropogenic VOCs are sparse. In Europe there is evidence that reductions were significant, comparable to NOx (see Lamprecht et al., ACP, 2020: doi: 10.5194/acp-21-3091-2021).

Line 168: Comment on: “the model is generally capable of reproducing the experimental data”: Looking at the ozone profile, it does not seem that the 3D model has a very robust predictive capability for ozone. In fact, Figure 2b shows a model offset between 10-20 ppbv for ozone concentrations around 50 – 60 ppbv (e.g. mid – troposphere), which is significant for ozone! For example regional AQ models typically reproduce tropospheric ozone within 5 ppbv when ozone concentrations are around 60 ppbv (e.g. Im et al., Atm. Environ., 2015; doi: 10.1016/j.atmosenv.2015.02.034 ). These AQ CTMS show biases at the high (e.g. >90ppbv) and low end (<30 ppbv), but not so much in the range of 50-60 ppbv. Is there an explanation for the large model bias in the mid to upper troposphere? Both HO2 and CH3O2 are overpredicted in the mid troposphere – it appears that additional ROx losses are missing in the model, or that the simplified experimental analysis for CH3O2 has limitations. Could the representation of clouds and liquid chemistry be a limitation, or are additional losses of RO2 x RO2/HO2 type reactions missing?

There is also indirect evidence of uncertain (e.g. VOC?)  chemistry. For example, in Figure 2 g a comparison between modelled and experimental CH3O2 concentration is shown. The mean difference from 3 km upward can be as large as a factor of 2! For comparison measurements and modelling by Ridley et al. (JGR, 1992, doi: 10.1029/91JD02287) showed that in the remote marine free troposphere, where CH4 and CO dominate, peroxy radicals estimated from a photo stationary state assumption can be largely reconciled with a photochemical box model.

Section 3.2

Line 212. Considering that the presented analysis in this section is exclusively based on the ECHAM/MESSy model scenarios here, why leave this statement as a possibility or “possible explanation”?  It should be pretty straight forward to get the emissions data from the model and compare the model projected changes quantitatively. E.g. how much have NOx emissions in the model then changed between 2003 and 2021?

Section 3.3

While I understand that alpha could be a semiquantitative experimental measure for investigating the chemical regime of ozone production, I wonder whether this section represents a little bit of a circular argument: since the analysis is largely based on the output of a chemical Earth system model (ECHAM/MESSy) anyway, why not also use established methods (e.g. Kleinman et al., GRL, 1997; doi: 10.1029/97GL02279) to investigate net P(O3) changes prior and post lockdown. From Figure 5 b and 5c, the relationship between alpha and NO for the individual campaigns does not seem to be dramatically different. From Figure 5a the difference between BLUESKY and BLUESKY-NL seems to be smaller than the uncertainty of both. So how robust are the findings? For example, if lockdown induced changes in anthropogenic VOC and NOx are proportional, one would expect to move sideways down along ozone isopleths. In this context it would be interesting to calculate the OH reactivity from the model. From the presented results and analysis, I have the impression that it is assumed to be dominated by CH4 and CO. While perhaps plausible in the upper remote atmosphere, it is hard to believe that VOCs wouldn’t play a significant role in the lower 3-5 km. Even in the remote (marine/coastal) atmosphere (e.g. Mao et al., ACP, 2009. doi: 10.5194/acp-9-163-2009) observations show that the VOC reactivity accounts for 20%. Aircraft studies have shown that models significantly underpredict VOC reactivity above North America (e.g. Chen et al., ACP, 2019: doi: 10.5194/acp-19-9097-2019), and that the VOC reactivity likely accounts for more than 40-50% in the FT over continental areas. This has been shown by many aircraft studies (e.g. Schroeder et al., Elementa, 2020: doi:  10.1525/elementa.400; Hu et al., JGR, 2014: doi: 10.1002/2014JD022627).

In summary: Putting the analysis more into context of the above mentioned literature and performing some sensitivity analysis on VOC reactivity would help clarify uncertainties that are associated with the main findings on ozone sensitivity.

Minor comments:

Line 30: Reaction R1 has already been described by Leighton

Line 35: This has already been shown by many studies in the 70ies and early 80ies (e.g. Calvert and Stockwell, Can. J. Chem., 61, 1983).

Line 44: what is meant by share here? the authors seem to refer to a fraction or a ratio in eq. (1)

Line 300: The authors sometimes put units in brackets e.g. [ppbv], but for alpha, which is a relative quantity, an empty bracket [ ] seems somewhat arbitrary.

This paper reports on new and exciting data from the BLUESKY experiment conducted ,during the early months of the COVID-19 pandemic.  The paper is well written and the figures are clear and appropriate.  In general I think that the aim of the paper is quite novel, and that the results could be useful for the research community.  However, I have reservations about the analysis because the BLUESKY ozone data in 2020 are not lower than the ozone values in the earlier campaigns. In contrast, three new studies show ozone above Europe was anomalously low in 2020.  This discrepancy needs to be reconciled before I can recommend the paper for publication.

Major comments:

1) Three recent papers have shown a clear decrease of free tropospheric ozone above Europe during 2020, in association with the COVID-19 economic downturn.  Two of these papers have been cited [Steinbrecht et al., 2021; Clark et al., 2021].  The third paper is by Chang et al., 2022, and it will appear any day as an accepted paper in AGU Advances (it will be posted here: https://agupubs.onlinelibrary.wiley.com/toc/2576604x/0/ja ).

Given that 2020 was an anomalously low year for ozone, it is very puzzling as to why the BLUESKY ozone observations are not lower than the UTOPIHAN or HOOVER data.  One possibility is that the sample size of these datasets is too low to provide an accurate estimate of monthly or seasonal mean ozone.  Three papers have looked at the sample size necessary to quantify monthly mean ozone above Europe and determined that 12-20 profiles are necessary [Logan et al., 1999; Saunois et al., 2012; Chang et al., 2020].  Given that the IAGOS program has dozens of profiles per month from Frankfurt, you could compare your monthly mean profiles to those from IAGOS.  The IAGOS monthly means will be accurate due to the high number of profiles and you can then determine if the aircraft campaign data are biased high or low.

2) Given that the three studies mentioned above report anomalously low ozone above Europe in 2020, we can conclude that net ozone production was below average in 2020, which matches the findings of Miyazaki et al., 2021.  However, your conclusion seems to be that net ozone production was not unusual in 2020. How can you reconcile these different conclusions?

3) Bouarar et al., 2021 concluded:

“Zonally averaged ozone in the free troposphere during Northern Hemisphere spring and summer is found to be 5%–15% lower than 19-yr climatological values, in good agreement with observations. About one third of this anomaly is attributed to the reduction scenario of air traffic during the pandemic”.  As conclusion that the reduction of aircraft emissions impacted ozone in 2020 has already been published, you should specifically mention this finding in your paper.  It would also be helpful to explain how aircraft emissions have strongly increased over the past 20 years [Lee et al., 2021].

4) I don’t agree with this statement in the Conclusions:

“We encourage future studies to investigate governing chemistry in the upper troposphere, a topic which has not received much attention in literature so far”

I know of many measurement and modelling studies of the chemistry of the upper troposphere, and a few that immediately come to mind are:  Barth et al., 2021; Brunner et al., 1998,2001; Cooper e t al., 2006; Huntrieser et al., 2002; Li et al., 2001,2005; Ridley et al., 1994.

If your comment is meant to refer to a specific chemical process in the upper troposphere, please make that point clear.

Minor comments:

First line of the Abstract:   lead should be led

Line 28-29

This line mentions ozone impacts on humans, animals and plants

“NOx directly impacts the production of tropospheric ozone (O3) which is a hazard to human, animal

and plant health (Nuvolone et al., 2018).”

However, the reference only deals with impacts on humans.  A good reference for the impact of ozone on plants is Mills et al., 2018.  I do not know of any authoritative papers that report ozone impacts on animals. If the authors know of such a paper they need to cite it, otherwise, impacts on animals should not be stated as there seems to be no convincing evidence.

Line 76

When reviewing studies that indicate ozone reduction in the free troposphere, you should also mention two recent studies that show ozone reductions at high elevation sites within the European boundary: Cristofanelli et al., 2021; WMO Air Quality and Climate Bulletin, 2021.

Line 166

“what” should be “which”

Line 192

“trend” is not the right word as it refers to a change with time. Would “gradient” work better?

References

Barth, M.C., Cantrell, C.A., Brune, W.H., Rutledge, S.A., Crawford, J.H., Huntrieser, H., Carey, L.D., MacGorman, D., Weisman, M., Pickering, K.E. and Bruning, E., 2015. The deep convective clouds and chemistry (DC3) field campaign. Bulletin of the American Meteorological Society, 96(8), pp.1281-1309.

Brunner, D., J. Staehelin, and D. Jeker (1998), Large-scale nitrogen oxide plumes in the tropopause region and implications for ozone, Science, 282, 1305–1309, doi:10.1126/science.282.5392.1305.

Brunner, D., J. Staehelin, D. Jeker, and H. Wernli (2001), Nitrogen oxides and ozone in the tropopause region of the Northern Hemisphere: Measurements from commercial aircraft in 1995/1996 and 1997, J. Geophys. Res., 106, 27,673– 27,699, doi:10.1029/2001JD900239.

Chang, K.-L., et al. (2020), Statistical regularization for trend detection: An integrated approach for detecting long-term trends from sparse tropospheric ozone profiles, Atmos. Chem. Phys., 20, 9915–9938, https://doi.org/10.5194/acp-20-9915-2020

Chang, K.-L., O. R. Cooper, A. Gaudel, M. Allaart, G. Ancellet, H. Clark, S. Godin-Beekmann, T. Leblanc, R. Van Malderen, P. Nédélec, I. Petropavlovskikh, W. Steinbrecht, R. Stübi, D. W. Tarasick, C. Torres (2022), Impact of the COVID-19 economic downturn on tropospheric ozone trends: an uncertainty weighted data synthesis for quantifying regional anomalies above western North America and Europe, AGU Advances, in press:  https://agupubs.onlinelibrary.wiley.com/journal/2576604x

Cooper, O. R., et al. (2006), Large upper tropospheric ozone enhancements above midlatitude North America during summer: In situ evidence from the IONS and MOZAIC ozone measurement network, J. Geophys. Res., 111, D24S05, doi:10.1029/2006JD007306.

Cristofanelli, P.; Arduni, J.; Serva, F. et al. Negative Ozone Anomalies at a High Mountain Site in Northern Italy during 2020: A Possible Role of COVID-19 Lockdowns? Environ. Res. Lett. 2021, 16 (7), 074029. https://doi.org/10.1088/1748-9326/ac0b6a

Huntrieser, H., Feigl, C., Schlager, H., Schröder, F., Gerbig, C., Van Velthoven, P., Flatøy, F., Théry, C., Petzold, A., Höller, H. and Schumann, U., 2002. Airborne measurements of NOx, tracer species, and small particles during the European Lightning Nitrogen Oxides Experiment. Journal of Geophysical Research: Atmospheres, 107(D11), pp.ACH-5.

Lee, D.S., Fahey, D.W., Skowron, A., Allen, M.R., Burkhardt, U., Chen, Q., Doherty, S.J., Freeman, S., Forster, P.M., Fuglestvedt, J. and Gettelman, A., 2021. The contribution of global aviation to anthropogenic climate forcing for 2000 to 2018. Atmospheric Environment, 244, p.117834.

Li, Q., D. J. Jacob, R. Park, Y. Wang, C. L. Heald, R. Hudman, R. M. Yantosca, R. V. Martin, and M. Evans (2005), North American pollution outflow and the trapping of convectively lifted pollution by upper-level

anticyclone, J. Geophys. Res., 110, D10301, doi:10.1029/2004JD005039.

Li, Q., Jacob, D.J., Logan, J.A., Bey, I., Yantosca, R.M., Liu, H., Martin, R.V., Fiore, A.M., Field, B.D., Duncan, B.N. and Thouret, V., 2001. A tropospheric ozone maximum over the Middle East. Geophysical Research Letters, 28(17), pp.3235-3238.

Logan, J. A.: An analysis of ozonesonde data for the troposphere: Recommendations for testing 3-D models and development of a gridded climatology for tropospheric ozone, J. Geophys. Res.-

Atmos., 104, 16115–16149, 1999.

Mills, G., et al. (2018), Tropospheric Ozone Assessment Report: Present-day tropospheric ozone distribution and trends relevant to vegetation, Elem. Sci. Anth., 6(1):47, DOI: https://doi.org/10.1525/elementa.302

Miyazaki, K., Bowman, K., Sekiya, T., Takigawa, M., Neu, J. L., Sudo, K., Eskes, H. (2021). Global tropospheric ozone responses to reduced NOx emissions linked to the COVID-19 worldwide lockdowns. Sci. Ad., 7 (24), eabf7460. doi: 10.1126/sciadv.abf7460

Ridley, B. A., J. G.Walega, J. E. Dye, and F. E. Grahek (1994), Distributions of NO to the upper troposphere, J. Geophys. Res., 109, D17305, doi:10.1029/2004JD004769.

Saunois, M., Emmons, L., Lamarque, J.-F., Tilmes, S., Wespes, C., Thouret, V., and Schultz, M.: Impact of sampling frequency in the analysis of tropospheric ozone observations, Atmos. Chem. Phys., 12, 6757–6773, https://doi.org/10.5194/acp-12-6757-2012, 2012.

WMO Air Quality and Climate Bulletin, No. 1, September 3, 2021; Editors:  O. R. Cooper, R. S. Sokhi, J. M. Nicely, G. Carmichael, A. Darmenov, P. Laj and J. Liggio; a publication of the World Meteorological Organization, https://library.wmo.int/index.php?lvl=notice_display&id=21942#.YTIzN99MG70