Overall Comment. The paper presents new observations of ozone over the East Antarctic Bharati station and interprets them in terms of stratospheric-tropospheric transport using a chemistry-climate model and trajectories back from soundings.  It will be of interest to readers who expect to see the data online.  Main recommendation: post the ozonesonde data- have you sent this to woudc.org and registered the site with them?  Your institutional link should be a primary location but the ozone profile user community will look for the data at woudc.org. Thank you.

Minor Comments:

• Although this reference is not new, it includes examples of chemical processes at the snow-ice interface that could be relevant to your study: Biogeochemical Cycles and Ice Cores, NATO ASI Series I-30, ed. R. J. Delmas. ISBN 3642511740; ISBN-13 978-3642511745.   Did you consider gases interacting with ozone that may come from snow-pack during the summer?
• Lines 280-286 – Vague. How would blizzards, etc, affect the observations? You imply model improvements need to be made to give a more accurate simulation. Be more specific.
• Lines 505-510. The R for Syowa disagreement suggests that the model is not doing well in winter at al.  What does this mean?  Reasons?

Girach et al. analyze surface ozone measurements at the Indian East Antarctic Bharati station in comparison with simulations from the chemistry climate model EMAC. The model is used to discriminate the role of subsidence from the stratosphere versus photochemical production in the troposphere or at surface level.

My main criticism is that the model description is not clear enough if or how chemical processes at the snow surface are included, in particular NOx emissions from the snow pack halogen/bromine chemistry, and the dry deposition of ozone. Even if it is argued that these processes may not be relevant for this particular study, at least it should be made clearer in Secton 2.2 which chemical processes are considered by the present EMAC simulations. Relevant EMAC sub-models seem to be available for polar bromine chemistry (Falk and Sinnhuber, 2018, https://doi.org/10.5194/gmd-11-1115-2018) and this latter study also investigated the uncertainty in ozone dry deposition velocities (as did other studies).

Overall the manuscript is well written and will be an interesting contribution to the literature, investigating the processes that contribute to ozone variability and trends in a data sparse region. I recommend publication in Atmos. Chem. Phys. after the following comments are taken into account.

Specific comments:

l.217: The statement on the tropopause fold occurrence frequency is somewhat disconnected and it is not clear how the conclusion can be made that the stratosphere to troposphere ozone flux is dominated by tropopause folds in contrast to slow subsidence through the tropopause.

l.250: O3s and O3t are correlated “due to mixing during the transport from the tropopause”. This statement confuses me. So does that mean O3s and O3t do not really represent the stratospheric and tropospheric contributions any more, but rather a mixture of the two? How useful are they then as a diagnostic??

l.252: “Strong local O3 production (e.g., through NOx from snow)”: again, is O3 production through NOx from snow included in the EMAC simulations? If not, is this some kind of circular reasoning? If it is included, would be good to give a few more details.

l.285: “to further improve the model in future studies”: how? Can you give some hints what may need to be improved?

Fig. 6b/c: The modelled net chemical tendencies (up to around 15 pmol/mol/h) are 1 order of magnitude smaller than the mean observed O3 tendencies (on average around 0.2 nmol/mol/h during morning and noon). Are the observed O3 tendencies in 6c not statistically significant? Or are there removal processes missing?

l.423: can you give us some idea what the “other” ozone production includes?

Minor comments:

l.65: “increasing trend (<0.2 nmol/mol/y)”: the number refers to a trend, not a trend increase. So either give some numbers how the trend is increasing or delete the word “increasing” in this context. Moreover, the “<” means this is an upper limit for the trend; better give a lower limit if available.

l.209: I suppose “Summit” is on Greenland? Why is this included here? If Summit should be included, please include lat/lon and/or geographic reference.

Fig.1 Caption: suggest to include explicitly the time period shown

Fig.2 Caption: suggest to mention that O3s and O3t are on a different scale than O3. (It took me a while before I realized that, couldn’t make sense of it before)