This study presents the analysis of observed high-latitude tidal and trace gas (ozone, water vapor) variability in response to SSWs, in combination with modeled trace gas and heating rate variations. The authors seek to leverage the simulated heating rates to quantify the impact of ozone and water vapor radiative perturbations on tidal variations in the MLT. By highlighting the role of trace gases, the authors aim to present the first comprehensive attempt to explain mesospheric tidal variability in the polar region. While the interpretation of the results is built upon a well researched literature framework, the connection between known mechanisms of tidal variability and the results of the current work is done qualitatively.

Since the main aim of the paper is to extend previously published results on ozone and water vapor coupling to mesospheric tidal variability, by specifically quantifying the contribution arising from polar trace gas variations, my main concern is that, in my view, this relationship is not in fact quantified. The paper remains at the stage where observations are argued to be explained by a possible mixture of dynamical and radiative effects, falling short of an actual quantification of the isolated radiative effects that are central to this work. I am further concerned that the relationship between SSW-driven ozone perturbations (and the corresponding radiative effects) and tidal variations might be hardly quantifiable, due to the lack of sunlight at these latitudes (upwards of 65 degrees North) during wintertime when SSWs occur. While the authors demonstrate that stratospheric and mesospheric altitudes receive some sunlight at these latitudes (even though the exact dates of the individual SSWs are not specified in the text; with some presumably occurring during the polar night), the modeled 0.5K/day perturbation is undoubtedly of very low intensity. While some context is provided for the modeled 0.5 K/day heating rates, the actual tidal forcing components of these rates are not discussed.

To support the author’s conclusion that this work provides a deeper understanding of the mechanisms driving tidal variability during SSWs, I therefore think that as a first step, the tidal forcing terms should be much more rigorously quantified. But even then, I would be surprised if the tidal forcing will be significant enough to result in an observable perturbation, given also the large number of other mechanisms and forcing terms involved in driving tidal variability during SSWs.

Given the tools used by the authors, possible suggestions for quantifying the tidal forcing perturbations could be, for example, to repeat the SD-WACCM-X experiments with specified ozone concentrations (preferably only past 65 degrees North), and/or to calculate Hough mode tidal forcing terms (or even simply 24, 12, and 8 hour tidal harmonics at the radar sites) from the 3-hourly SD-WACCM-X heating rates. 

More general comments are given below:

The abstract mentions that the water vapor perturbations have a negligibly small impact on tidal changes. As a suggestion, I would therefore consider changing the title to “New insights into polar ozone and water vapor, and the connection between ozone radiative effects and tides in the mesosphere-lower thermosphere during major sudden stratospheric warming events”. I would also consider changing line 19 to read “...transport of radiatively active ozone is important for explaining the observed tidal variability”, rather than “transport of radiatively active gases is important…”.

L6: ‘polar latitude’ to ‘polar latitudes’.

L10: ‘TDT tide’ to ‘TDT amplitudes’.

L50-51: I would suggest adding information that the intensified tides and quasi-two-day wave amplitudes observed by Lima et al. (2012) were observed in the low-latitude summer hemisphere.

L76-79: The authors note that this study, for the first time, quantifies the impact of ozone and water vapor responses on tidal variations in the MLT. Do the authors mean to say the ozone and water vapor responses specifically at polar latitudes? The impact of ozone perturbations on MLT tidal variations has been investigated in numerous other studies, as also referred to in the introduction. It is also not clear to me how the paper presents a quantification of the total radiative forcing during SSW events; maybe total radiative forcing above 65 degrees North is meant, based on the results showing averaged data between 65-90 degrees North?

L144: Perhaps it is enough to mention that SD-WACCM-X can be run in a fully coupled mode. Currently it is not clear whether the model is run with an active or prescribed ocean, which in a way distracts, since this is not strictly necessary information for understanding the results of this work.

L152: I would suggest rephrasing “photochemical rate constants” to “photochemical absorption and quantum-yield data”, or alternatively “photochemical molecular data” since currently it could be interpreted as if the photochemical reaction rate coefficients (J-values) are kept constant, but clearly these vary as a function of solar zenith angle and trace gas concentrations.

L154: Could it be specified which years exactly the climatological SD-WACCM-X simulations span? Given that short-wave radiation absorption is central to the paper, could the (E)UV absorption scheme and relevant chemical species from WACCM-X also briefly be discussed?

L169-L170: A table showing the central SSW dates between 2004-2022 would be highly beneficial, given that insolation can be very different between an event occurring on the 24th of March (in 2010) or 5th of January (in 2004). Could it also be specified which model from the cited NOAA page is used as reference (even though I would think this is MERRA-2 based on its relation to SD-WACCM-X)? It would also be helpful if the exact SSW onset criterion is specified in the text, given the large variety of definitions used in literature. 

L174-177: To my knowledge, not all major SSW events are associated with an Elevated Stratopause (ES). Looking at the NOAA database, there are 8 major SSWs between 2004-2013. However, Limpasuvan et al. (2019) identified only 5 ES SSWs during this time period (their Figure 1). Please discuss the notion of not all SSWs having an ES, and how this affects the interpretation of your results, as ES SSW characteristics are assumed to be present as a general feature in the modelled and measured composite SSW response later on in the text.

L180: The link between planetary-wave activity and the observed oscillations in the meridional winds is unclear to me. While I understand that planetary wave-mean flow interactions can induce low-frequency oscillations in the mean meridional winds, the analysis described in Section 2.1 states that the time-frequency analysis also includes longer period waves such as stationary planetary waves, in addition to a mean wind. How is it possible to differentiate between low frequency (quasi-stationary) planetary waves and mean wind oscillations at a single station without additional longitudinal information?

L188: At this point naming the station coordinates becomes a bit repetitive. My suggestion would be that here “the three radars” is simply enough.

L192: The interpretation of the %-changes in tidal amplitudes and mean winds would benefit from a comparison to climatological values. Could the climatological values between for example January and March be included as figures in the appendix? And similarly for the long- and short-wave heating rates? Given the spread in SSW onset dates (4th of January to 24th of March), I would expect there to be a considerable spread in event-to-event QRS rates and the associated deviations from climatology, given the rapid fall off of insolation during winter. It would probably also be helpful to discuss the event-to-event variability in QRS in the paper, based in part on the climatological figures.

L195: Possible reasons for the observed TDT variations are suggested here. Would this be more appropriate for the discussion section? Alternatively, I would suggest moving certain discussion points, for example that the TDT may be contaminated by gravity waves in the analysis technique, to this section. With this in mind, could the authors comment on why the TDT enhancement 10 to 20 days after the SSW onset is so sporadic? For example, at the Tromsø site, the meridional anomaly is around + 2 m/s on day 30, then falls to zero, and then returns to + 2 m/s on day 40. How does this variability fit in with the kind of fluctuations that could be expected from GW contamination?

L201: Please be more specific about how exactly the TDT variability observed in the current work aligns with previous studies that used GCMs and satellite measurements to discuss the solar heating, nonlinear interactions, and gravity wave-tide interaction excitation mechanisms for the TDT. This also ties in with the above comment. 

L200: Do the authors here mean to say “TDT amplitude anomaly”?

L213: Personally I do not see a clear sign of the STD showing an enhancement 20-50 days after the SSW onset (albeit weaker than during SSW onset time), at least not in the data presented in the current work. In the current work the STD shows considerable variability also during times without SSWs, while there is no commonly identifiable pattern between the three meteor radar stations between days 20-50. If anything, I would argue that only Sodankylä shows a local maximum between days 40-50. 

L234: It is not clear to me what Figure 5 shows. From the context I would guess that these are daily averaged 3-hourly heating rates? Or are they amplitudes of the 24 hr variations in the (3-hourly) modeled heating rates? This is crucial information for understanding discussion points later on. 

L254: This is largely a repeat from an earlier comment, but I fail to see a clear SDT enhancement at the three stations 20-50 days after SSW onset. It is therefore difficult to connect the modeled QRS rates to SDT amplitude anomalies 20-50 days after onset. As a suggestion, could a time-series of, for example, 90 km altitude SDT amplitudes and 50 km altitude QRS rates be shown in a single figure? This may help to demonstrate a more clear relationship between the two. 

L255: I do not quite follow the line of reasoning where 1) the findings from Siddiqui et al. (2019) and Limpasuvan et al. (2016) that SSWs are followed by rapid increase of ozone between 20S to 40N arising from equatorial upwelling and cooling, and a decrease poleward of 40N, and 2) the subsequent ozone enhancement at mid to high latitude as shown in Figure 6 happening immediately after SSW onset. Based on the first point, I would expect a rapid decrease in ozone centered on the SSW onset date in Figure 6? 

L256: I would suggest adding that these heating rates are for wintertime. I would further expect these heating rates to be quite different between January and March SSWs, given the high latitude of the stations, and considering that heating rates approach 14 K/day during summer (Brasseur and Solomon, 2005, Figure 4.25). As mentioned above, I think it would therefore be highly beneficial if climatological heating rates between, say, January and March could be added to the appendix. 

L257: It is argued that the QRS change is mostly due to ozone increases following the SSW onsets. The QRS anomaly falls roughly between 0.1 to 1 hPa (Figure 5), while the ozone anomalies extend between roughly 50 to 0.1 hPa in two largely separate patches (Figure 6). Why wouldn’t the ozone anomaly below 1 hPa, i.e. the bottom patch, contribute to the QRS anomaly? Is this because the stratosphere does not receive any sunlight at these latitudes during winter? A comment at this point in the manuscript would be beneficial. The lack of QRS perturbation between 100-10 hPa seems to conflict with the stratosphere receiving sunlight as argued for in section 3.3.

L286: Could the double-layer structure of the ozone be discussed or clarified in more detail? Does this refer to the anomaly in Figure 6 appearing to show a two-layer structure? I would think that a two-layered anomaly would not necessarily imply a double-layered structure of the underlying layer. Could the upper anomaly (centered on 1 hPa, or 50 km) simply be an extension of the stratospheric ozone layer?

L286-300: It is hard to imagine a 0.5 K/day heating rate (if these indeed would amount to 0.5 K/day) to provide significant tidal wave energy. In my view, isolated model experiments and a more detailed discussion of the tidal forcing terms would be a necessary step here. As a first estimate, what is the amplitude of the 12 hr component in the heating rates?

The stratospheric diurnal tide is further a largely vertically trapped mode, so it is unclear to me how the amplitudes at stratospheric altitudes (40-50 km) translates to amplitudes at meteor radar altitudes (80-100 km). The superposition of two 24 h waves will also always result in a 24 h wave, no matter how in or out of phase they are (so long as their frequencies are both 24 h). So it is also unclear to me why the superposition of two diurnal waves may effectively amplify the SDT at meteor radar altitudes? Regardless, I think the impact of travel time and vertical distance should be explained in more detail.

L298: I can’t find reference to the (in-situ) diurnal heating rates causing pronounced diurnal tides at the latitude and altitudes relevant to the current work based on the cited work by Schranz et al. (2018), which appears to discuss only the diurnal cycle in ozone and not the winds. Could this be clarified?

L310: Here the observed SDT response is primarily attributed to changes in zonal mean wind and ozone heating at mid-to-low latitudes, even though the contributions of these effects are not quantified, and these are therefore difficult to place into context with the ozone mechanism described in the paper. The double-layer ozone structure is further argued to contribute to the immediate STD response on line 315, while on line 312 it is argued to likely contribute to the observed STD variability during the recovery phase, and not during the onset phase. This seems contradictory. 

L309-L324: I struggle to see the connection between the discussion points in this paragraph and the placement of the results within the literature. I think this largely stems from the contributions of the different mechanisms (propagation conditions, mid-to-low latitude ozone forcing) not being quantified in the context of the observational data, even though observed characteristics are attributed to these mechanisms. When the aim of the paper is to quantify the contribution of polar trace gas perturbations, I think a more careful quantification or discussion of the other effects is also warranted, given that the net observed tidal response is shaped by the complex interplay of all the different mechanisms. Further, as mentioned above, while the modeled 0.5 K/day heating rates fall within the range of stratospheric diurnal temperature variations, the actual diurnal components of the heating rates are not discussed. 

Please check the spelling of Tromsø in the figure sub-titles (sometimes spelled as Tromose).

References

Sridharan, S., S. Sathishkumar, and S. Gurubaran. "Variabilities of mesospheric tides during sudden stratospheric warming events of 2006 and 2009 and their relationship with ozone and water vapour." Journal of atmospheric and solar-terrestrial physics 78 (2012): 108-115.

Limpasuvan, Varavut, et al. "On the composite response of the MLT to major sudden stratospheric warming events with elevated stratopause." Journal of Geophysical Research: Atmospheres 121.9 (2016): 4518-4537.

Brasseur, G. P. and Solomon, S.: Aeronomy of the middle atmosphere: Chemistry and physics of the stratosphere and mesosphere, vol. 32, Springer Science & Business Media, 2005

Schranz, F., Fernandez, S., Kämpfer, N., and Palm, M.: Diurnal variation in middle-atmospheric ozone observed by ground-based microwave radiometry at Ny-Ålesund over 1 year, Atmos. Chem. Phys., 18, 4113–4130, https://doi.org/10.5194/acp-18-4113-2018, 2018.

Review of Shi et al

The study by Shi et al utilized observational data of winds, ozone and water vapor at high northern latitudes to analyze the response of tidal wind amplitudes to sudden stratospheric warmings (SSWs), and ask the question whether anomalies in radiative active trace gas abundances may contribute to the anomalies in tides. Global model data are used to deduce anomalies in radiative heating rates before, during and after SSWs. The study presents interesting signals in tidal amplitudes of the diurnal, semi-diurnal and terdiurnal tides at three different latitudes. Furthermore, trace gas anomalies (water vapour and ozone) associated with SSWs are shown to be very consistent between satellite data (MLS), global model simulations (WACCM-X) and local measurements in Svalbard, which is a very encouraging result. 

The consistent and comprehensive quantification of tidal and trace gas anomalies around SSWs from observational data is a valuable set of analysis worth publication. The study further puts forward the suggestion that the trace gas anomalies are important in "explaining the observed tidal variability during SSW events" (abstract, line 19-20). While I agree that the modification of tides by ozone anomalies is a plausible mechanism, I have to disagree that the pieces of analysis shown in the study provide any quantification of this radiative effect, as detailed below. It is a valid and interesting point to discuss the possible impact of the trace gas anomalies (and associated heating anomalies diagnosed from the model) on the tidal anomalies, but the authors should present those points as discussion of possible (!) effects, rather than stating that a comprehensive explanation of tidal variability is obtained in the study. Furthermore, there are a number of corrections necessary in to improve the presentation of the results, and at a few places I found the description of the results inconsistent. Overall, I recommenced that the authors revise their manuscript majorly, focusing the study on the quantification of SSW signatures in tides and tracers.

Major comments:

--------------------------

1) Quantification of role of trace gas anomalies to force tidal anomalies

The authors state that they "presents the first comprehensive attempt to explain mesospheric tidal variability ..." (line 306, similar in abstract line 19-22). As stated above, I generally agree that the study presents an interesting and comprehensive quantification of tidal variability and co-located tracer variability from observational data, and it is valid to discuss whether there could be possible effect of tracer anomalies on tidal variability. However, I have to disagree that the study is a comprehensive attempt to explain the tidal variability, for which an actual quantification of the role of trace gas anomalies for the tidal generation would be necessary. 

While the presentation of the radiative heating rates from model data help to infer whether there might be an effect at all from the tracer anomalies, this is not sufficient to conclude on whether the tracer anomalies actually play a role to cause the tidal anomalies. For this, as least two more steps would be necessary: a) quantify how much of the radiative anomaly is actually due to the tracer anomalies; b) quantify how much the anomalous radiative heating/cooling contributes to tidal generation. The former could be done by offline radiative calculations. A full quantification would, as far as I can see, only be possible by conducting model simulations which either include or discard the tracer anomalies. As of now, the statements made in the paper on the role of tracer anomalies for tidal variability are at most based on scaling arguments (e.g., role of tracer anomalies for heating, see comment on lines 257-259; argument about similar strength of relative anomalies in line 365-366), and many very speculative statements are made in particular on the relative role of anomalous heating/cooling via tracer anomalies versus anomalous propagation due to changed background winds (e.g., on page 15, see individual comments below on line 257-259, line 284, 286). 

I understand that conducting simulations for the actual quantification would go beyond the scope of the paper. Therefore, I suggest that the authors focus on the quantification of the tidal and tracer anomalies from observations, which by itself composes an interesting set of analysis. This can be complemented by a discussion of possible effects of the tracer anomalies on tides, but making sure to emphasize the uncertainties and the speculative nature of some of the arguments/hypothesis put forward.

2) Inconsistencies in the discussion on SDT

Much of the discussion on the effects of ozone anomalies on tides focuses on STD, which was shown to be enhanced just around the central date of SSWs. I have to admit that I got confused about the timing of when this link is suggested to act. At places, it is stated that the STD enhancement is linked to the persistent ozone (and associated QRS) anomaly over several weeks after the SSW (e.g., in line 253), but this is inconsistent with the timing of the STD anomaly just around the SSW event. At other places, the STD enhancement is stated to occur at the SSW onset, and that it is "... attributed to zonal wind changes and ozone heating at mid- to low latitudes" (line 311). Firstly, there is no actual attribution presented in this study (see major comment 1 above), and secondly ozone heating at low- to mid-latitudes is not shown here, so this confuses me even further. Please ensure to clarify better the proposed link between ozone anomalies and STD anomalies (see also individual comments below on line 253, 311, 315, 329)

3) Methodological issues and presentation of results

- Some methodological clarifications are needed, e.g. stating more clearly which time periods are used for the different data sets and clarifying some details on how the tides were fitted and how the (relative?) anomalies were calculated. See individual comments below.

- Please add significance tests to all results to ensure that robust signals can be identified.

Individual comments:

------------------------------

Title: I suggest to revise the title; in its present form there might be a word missing ("the polar ozone and water vapour ANOMALIES (?)"), and the "radiative effects" in the middle of the title seems out of place. Also, is "polar" referring to everything, or just ozone (as it appears right now)? How about simplifying the title to something like "Response of tides in the polar MLT to SSW events, and their link to ozone and water vapour anomalies."

Abstract

line 7: "polar ozone and water vapour in linking mesospheric tidal variability...": what is "linking" referring to here? maybe you mean "driving", or "contributing to" (but those words would be too strong, see major comment 1). Please rephrase.

line 8-14: The description of anomalies in tidal amplitudes appears rather clearly written up until line 12, but misses to clarify which wind component the text refers to. Starting in line 11, a distinction is made between zonal and meridional wind components, making me wonder which component was refers to in the earlier sentences? Please clarify.

line 19-22: "The interaction between dynamic processes and the transport of radiatively active gases is important for explaining the observed tidal variability during SSW events": this is not backed up by the results, see major comment.

Section 1

- line 48: are the "modified zonal mean zonal winds" modifying tidal amplitudes via propagation of tidal waves? Please clarify and be more specific on the suggested mechanisms from literature.

- line 51: "heightened planetary wave activity" - change to "strengthened planetary wave activity"

- line 54: did the authors meant to distinguish between mixing ratio and ozone density within this sentence? 

- line 61: "ozone dynamics": consider changing to "ozone transport", not sure what "ozone dynamics" are.

- line 68: I'm not familiar with all the referenced studies, but at least the last three references (Oehrlein et al, de la Camera et al and Hong and Reichler) do NOT analyse tides in the MLT region; they analyze the impact of interactive ozone on stratospheric dynamics including planetary waves. Thus, they cannot serve as appropriate reference for the statements made here.

- line 76: QRS and QRL are defined in the abstract, but it would be good to repeat the definition here; also it might be more appropriate to just say radiative heating and cooling here rather than using the variable abbreviations.

- line 76: it would be good to mention at this point that comparison of water vapour and ozone anomalies from WACCM-X to observations are performed in a first step to ensure consistency with the observational data.

- line 83-84: this description of the structure of the paper is completely generic; it might as well be dropped, or better explain what is done in which section within the previous paragraph.

Section 2

- line 87: "at three different high latitudes"; consider changing to "at three stations in high latitudes"

- line 88: continuously in operation in which time period? Please specify.

- line 90: unclear what is meant by "the same wind retrieval algorithm" - the same over time, between the stations, or the same than in the reference? please clarify.

- line 93-94: "tides ... are obtained" - does this mean that the given equation is fitted to the data to obtain the tidal amplitudes? Please be more specific how the equation is used to extract tides. 

- line 98: where are the zonal mean variables obtained from for the analysis of the tides from the meteor radar? Or are they a result of the fitting?

- line 100: planetary waves (waves is missing)

- line 106 and line 117: is this the same "QPACK" software, since the reference is the same? If yes, please be consistent in the spelling; if not, clarify.

- line 132: please provide the approximate altitudes for the MLS data, given that meteor radar observations are provided in altitude coordinates. An interpolation to a consistent vertical coordinate would ease comparisons made later between tidal anomalies and trace gas anomalies.

- line 141: I suggest removing "A comprehensive numerical model.." (both unnecessary and incomplete sentence).

- line 143-144: "capable of being run ..." consider rephrasing, e.g. simply to "WACCM-X can be run with a coupled or prescribed ...". Generally, I suggest to rather focus on describing the configuration of WACCM-X that is used here, rather than listing possibilities and developments in the CAM/WACCM/WACCM-X framework (this list would be rather long)!

- line 150-153: I wouldn't agree that the developments described here are "recent" (going back to 2010), nor do I believe they are particularly relevant to be mentioned for the study here. 

- line 155: please rather mention the years of the simulation than saying they are done for the measurement period. Also, are those dedicated simulations performed for this study, or the ones available at: https://doi.org/10.26024/5b58-nc53. If not, what is different?

- line 159: here would be a good place to define the QRL and QRS heating rate variables in detail.

- General: the information on time period of available observations is largely missing in Section 2. Please add.

Section 3

- line 166: please add how many events the composites are based on (so the reader doesn't need to check the compendium for the given period). 

- line 174: "altitude of wind reversal": the reversal cannot be seen from the anomalies, consider adding the zero wind contour to the Figures.

- line 181: why would positive meridional wind anomalies be associated with reduced wave activity? Please elaborate.

- line 181: please specify in this sentence that the references named here (Dowdy et al and Koushik et al) refer to wind and wave response to SSW in the MLT. As the sentence reads now, it sounds like it refers to the dynamics of the wind reversal in the stratosphere (i.e., the SSW generation itself).

- line 191: I didn't quite understand how the anomalies in tidal amplitudes are calculates - it says here that they are a "relative change", but the units are m/s; also the phrase "taken as a mean time for the entire showcase period" is not clear to me. Furthermore, it would be very helpful to add a significance test to the anomalies in tidal amplitudes. This way one could infer which anomalies are statistically robust.

- line 194: percentage difference relative to the mean value (add "relative to")

- line 243: "This indicates...": it is not clear to me what is implied here: that the persistent ozone anomalies are arising from a persistent alteration of the thermal structure (via dynamics), or that the ozone anomalies themselves might give rise, or a least contribute to the persistence of temperature anomalies ?

- line 248: I agree ozone changes are dependent on altitude, but Fig. 6 does not provide information in variations with latitude - for the latitudes shown here, the ozone anomalies are remarkably similar.

- line 249ff: It s not clear to me which altitude, nor process is referred to here: I agree that there might be enhanced tropical upwelling during the SSW event due to pronounced wave driving in the stratosphere; However, enhanced upwelling would reduce tropical ozone in the lower and mid-stratosphere. Please specify which altitudes you refer to here. 

- line 251: "... a decrease poleward of 40°N. The ozone enhancement ... at mid to high latitudes is consistent...": This confuses me - how is the ozone decrease poleward of 40°N shown by previous work consistent with the enhancement of ozone diagnosed from the observations here? Do you refer to different timings (before versus after SSW)? Please clarify and rephrase.

- line 253, and general: "ozone increase.. and QRS therein": I agree that changes in ozone should primarily affect the shortwave rather than longwave radiation. Therefore, I am surprised why the authors decided to overlay the QRL contours on the ozone anomalies instead of the QRS fields, which would ease making the link between the tracer changes and the resulting radiative changes.

- line 253: Here, and at several places I was confused which anomalies are referred to when the "enhancement of STD" is discussed: according to Fig. 3, STD is mainly enhanced just around the SSW event (there is some sign of enhancement later, but very noisy signals including reductions, and I doubt this is significant). However, here the "ozone increase and QRS therein" in the "SSW recovery period (day 20-50)" are suggested to be coincident with the STD anomaly - can you clarify?

- line 257-259: I agree that it is plausible that the ozone anomalies are a main cause of the QRS anomalies, but it would of course be great if this would be quantified, which would be possible e.g. via offline radiation calculations. As is now, the only way to infer the importance of ozone for QRS is the alignment of anomalies, and the rough approximation saying that "mean heating rate by ozone at those latitudes is around 1 K/day". Even more important, the only argument by which the authors rule out the role of water vapour for radiative anomalies is saying that the 25% anomaly "translates to changes in cooling rate of about 0.05 K/day" (line 259). It is not clear to me what this number of 0.05 K/day is based on, and the same holds for the 1 K/day heating by ozone. Given the argument for the importance of  the tracers is almost completely based on those numbers, it needs to be made much clearer where they are taken from, for which altitude and latitude region they are valid, and in how far one can linearly scale the radiative impact with the tracer anomalies (as apparently done here).

- Figs. 6, 7 and 8: great to see the good agreement between the satellite, model and station data! Possibly this result could be emphasized more?

- line 268ff: I agree it is good to calculate the dynamical-driven temperature tendencies in order to compare them to the radative ones. However, for the purpose of the paper and in the following discussion it was not entirely clear to me what this analysis reveals beyond the fact that dynamical heating/cooling is generally balanced by radiative (in particular long-wave) cooling/heating?

- line 277: ".. the dynamic process drive the persistence of ozone anomalies...": I agree that this is likely the case, but the temperature tendencies do not necessarily help to explain the tracer anomalies - do you suggest it is the anomalous vertical circulation which drives both temperature as well as trace gas anomalies (the later via vertical advection)? Or do you suggest that the temperatures affect ozone via chemistry?

- line 281: "still remains in the Earth's shadow..": this would be a good place to provide details on the altitudes which are reached by sunlight as function of latitude. This is addressed by the schematic Fig. 11 later on; if kept, this schematic and the considerations with it should be mentioned here. However, I suggest replacing the schematic Fig. 11 by a figure showing the amount of sunlight in mid-winter as fct of latitude and altitude; this would help to make this point in a more quantitative manner. It could even be shown as a function of daytime to help to make the point on how / where ozone anomalies might affect tidal amplitudes.

- line 284: at this point, it appears very speculative to conclude that the redistribution of ozone contributes to tidal variability. I agree this might appear plausible, but the pure quantification of shortwave heating anomalies does not give any quantitative estimate on how strong the effect on tides might be.

- line 286: results on ozone double layer: I cannot, or barely, see how the double layer of ozone anomalies is reflected in "UV heating" (i.e. QRS) - do I assume correctly that the authors refer to the change of sign in QRS around 10 hPa, visible as change from light blue to light red in Fig. 5 (right)? If yes, it has to be clarified whether the heating in the lower layer is at all different from zero, as this is not apparent from the Figure.

- line 286: "these two diurnal tidal waves": do the authors suggest that the double layer of ozone, and thus possibly shortwave heating anomalies, cause two distinct forcing regions of tidal waves? To me, this result appears to be extremely speculative and needs to be justified much better.

- line 295: it would be great to give a specific altitude down to which the sun is above the horizon at the given latitudes instead of just mesosphere and stratosphere (see also comment above).

Section 4

- line 311ff: "explores the enhancement of SDT amplitudes at the onset of SSW ... attributed to zonal wind changes and ozone heating at mid- to low latitudes": This statement added to my confusion on what the authors suggest and present; Firstly, I do not see how the study attributed tidal variability in any way (see main comment); secondly ozone heating at low- to mid-latitudes is not shown in the paper, so this comes somewhat out of nowhere (unless I missed something?). It goes on discussing the STD anomalies during the recovery phase, which I cannot identify from the STD anomalies presented in Fig. 3 (see comment above).

- line 315ff: also here, I find it hard to follow the arguments by the authors; again a heating rate by ozone is mentioned (0.5K/day), but it is not clear what this value is based on or where it would be valid; further, it is said to be small (with respect to what?), but in the same range than tidal temperature amplitudes (so not small for tidal variability?)

- line 320: again, which time period is referred to for which STD is enhanced by 40%

- line 322: I agree that radiative effects are more important in the time following the SSW, but for what? For the mean temperature anomalies or the tidal amplitudes? (For the latter, this is not backed up by results).

- line 325ff: the discussion of the DT is much better to follow and to comprehend compared to the discussion on STD.

Section 5

- line 355: I disagree that a deeper understanding on mechanism is provided (see main comment 1)

- line 361-362: the altered background presumably affects propagation, but this not quantified here. Please clarify that this is a proposed mechanism, rather than a result of the study.