This is an ambitious paper with three goals, to show: 1) that the Australian wildfires of 2019 and early 2020 changed the character of the stratospheric aerosol incorporated into the 2020 austral polar vortex. 2) that these aerosol subsequently led to an enhanced abundance of PSCs, and 3)  that a new analysis of pairs of CALIOPI observations, of related spatial locations south of 80 degrees, from different orbits of the instrument, can be used to make conclusions about PSC formation pathways.

There is some interesting analysis preformed in support of these goals, but ultimately the authors struggle to carry through with enough detail on any one of these topics to make a convincing argument.  This deficiency increases as the paper moves through these three goals. The paper style is to describe each observational product with a short cursory analysis, reference and discussion of supporting literature, and then how this observation leads to the next issue to be tackled. A more normal style is to present a complete discussion of the background literature on the topics to be discussed, e.g. stratospheric aerosol and PSCs, in the introduction, not to scatter such presentations throughout the rest of the paper. Once the literature is reviewed then it doesn’t need to be continually re-cited. It will be obvious to the reader how the new analysis fits into the literature that was already reviewed in the introduction.

The paper cannot be accepted in its present form. Perhaps a paper tackling the first, or first and second, goals, and doing a complete job on that, has a possibility of acceptance. The third goal contains the germ of an interesting analysis idea, but requires significantly more work, much less generalization, and the need to ground any conclusions with a solid microphysical analysis. It is also incumbent for the authors to use the language of the PSC community and really reflective of the CALIOPI observations. Their current analysis is way too simplistic.

Here follow comments by line number initially tailored to help the authors with a better presentation, but then they evolved into questions/complaints/suggestions concerning the authors increasingly inability to present a complete picture and thorough analysis of a topic before moving on. Quotes from the paper are bracketed with ellipses.

24, 83 LNAT – This acronym is misleading. It suggests a liquid NAT particle which doesn’t exist. Stick with the nomenclature of Pitts et al., e.g. liquid-NAT mixtures. Then there is no confusion. Upon reading the abstract and seeing this acronym I was prepared to read a paper not conversant with the PSC literature or expecting to learn something I did not know. The latter is not the case. Don’t let the reader think the former. Don’t use this acronym.

 123 What usually occurs at ~-80˚? In the example shown the intersection is at -83˚ or so.

143 How are the vertical levels chosen and why only three? It seems there could be a number of vertical levels involved depending on the temperature structure.

144-145 Limiting the classification to only the most populated PSC type biases the results to only the dominant PSC kind. The different kinds of PSCs in any one air volume is not noise. It is not at all clear that this simplification provides any vital information about PSC formation pathways, rather it masks the reality.

According to Pitts et al. (2018) the CALIOPI data are averaged along track over 5, 15, 45, and possibly 135 km between 8.5 and 30 km to obtain a PSC classification. Here the authors downgrade these observations of 5 possible PSC classes (No Cloud, STS, NAT mixtures, enhanced NAT mixtures, ice) across an along track distance of 300 km, before and after the intersection point from the Nth scan to the Nth+2 scan, and at 250 m vertical resolution to just one PSC class. So roughly for each CALIOPI intersection point two measurements each containing at a minimum some 2000 possibilities (600/135 * (30-10)/0.25 * 5 = distance intervals * altitude intervals * PSC classes) is reduced to two classifications. This generalization is way too simplistic.

158-159. The intersection points occur over a range of latitudes < -80. State the range, not just that they are close to -80, which clearly from Fig. 1 they are not.

162 It is clear how the horizontal boundary is selected, but not how the vertical interval is selected. It was ambiguous before and remains so.

163-164 This is speculation about NAT rocks without any basis. In fact there is information in the first scan masked by the requirement that the most populated category, in this case NC, is used to characterize all the scans.

165-166 This is way too simplistic. What are the temperatures in the first and second scans? How much did the air move in this period? Where did it come from? Were there already vestiges of PSCs in the first scan? Are the first two scans related to each other in any way? Did the air move from the first to the second, or from the second to the first? The authors could obtain all of this information which is germane to each specific measurement.

166 … Similarly, all possible formation pathways of both ice and LNAT are retrieved during each successful intersection point… This has not been demonstrated or even discussed. Thus such a claim cannot be made.

168-175 This is idle speculation and should all be eliminated. Anyone studying PSCs already knows that temperature and gaseous constituents are the most important items. That is what the introduction is for.

178-179 … as described in Sect. 2.3, we accounted for changes in these parameters when retrieving the PSC formation pathways … This is not true. Where were these accounted for? Temperature was for example never mentioned.

215-216 … Since April 2020, a negative anomaly of kext has been observed at the latitude ~-80°, which is attributed to the nucleation of PSC on these aerosols (Zhu et al., 2018)…. It is not clear on the figure that this is the case. In April 2020 the colors are yellow and white, which straddle zero. Is this what the authors are pointing to? Also OMPS is struggling to reach -80 degrees in April, unlikely that the measurements extend beyond the middle of April. In any case if PSCs formed wouldn’t this increase the extinction? Is it Zhu et al. who attributed the nucleation of PSCs on these aerosols or the authors here?

216-217 … The kext increased again at high latitudes in October and November 2020 which is due to the re-release of the captured aerosols by the PSC, … This is pure speculation and makes no sense. If PSCs evaporate the particles get smaller and extinction decreases.

253-255 The water vapor decrease doesn’t really begin until late winter and then extends through spring to summer long after PSCs are a factor. The change in HNO3 is more indicative of being impacted by PSCs, but how is the 2020 winter different than others in this regard, since HNO3 is always observed to begin decreasing about this time at these altitudes? Linking this observation to the presence of brush fire smoke is too simplistic and is not verified by these observations.

Fig. 6 and discussion. What are the latitude/longitude bounds for the data shown? Why were these particular altitudes chosen? Were others looked at and these deemed to be representative or?

298 Fig. 7a) shows OMPS extinctions, not n2o5.

Fig. 7 What time period is represented by the blue lines and shading? What does the shading indicate?

317 … The near-simultaneous decrease in aerosol loading (as discussed in Sect. 3.2), … Where is this shown? It is not substantiated by Figs 3 or 4. In fact OMPS can provide little aerosol data at PSC relevant latitudes in austral winter.

333-335 These statement seem obvious. Any particle surface area promotes the chlorine activation reaction.

335-336 Wasn’t enhanced ozone loss observed in 2020? Why not reference that paper? The references chosen are really just background information on STS, not so applicable here and should have already been covered in the introduction.

337-339 Here ENAT has a very small anomaly in July and August between 0 and 1 (x1e6 km2) yet the standardized anomaly exceeds three standard deviations, which calls into question the usefulness of the standardized anomaly. Note that it matches or exceeds SA of STS which has a much higher anomaly.

338-339 …It is evident from Fig. 8 that the positive anomalies in the areal coverage of PSCs like ice, STS, and ENAT exceeded three standard deviations with respect to the background mean… Correct, and yet some of these differences are important and others are not. It is difficult to know how to interpret the standardized anomaly. If the standard deviation is small the standardized anomaly will be large.

343 Again there is no such thing as a liquid nitric acid trihydrate. It is an oxymoron. Hydrate implies a solid particle. Weren’t all these acronyms already defined anyway?

375-385 There are numerous problems with this discussion. First there is no nucleation barrier for STS it is a liquid solution droplet of water, hno3 and h2so4. Second all of these explanation have either already been repeated or should have appeared in the introduction and background on PSCs. Third the authors here use a very simplistic approach of attributing all of a particular CALIOPI observation to the most dominant particle in the observation. Thus NC may well contain PSC particles, they just don’t happen to be the dominant ones. Fourth do the changes in temperature make sense to convert NC into a liquid NAT mixture based on the HNO3 and H20 available? This can be calculated. Fifth the spread in the temperature data to convert from NC-LNAT is from -1 to 0.8 K for the interquartile range. So LNAT forms even though the temperature increases from a NC situation? This alone should give the authors pause.

390-480 Here the authors carry on with their discussion of the different PSC formation pathways, but without providing anything more than statistical broad brushes and without any quantification checks. For example, are the temperature changes consistent with the changes in hno3 concentrations for various transitions. The final conclusion seems to be.

…From the above discussions, it is clear that the majority of the LNAT (82 %) and ice (77 %) are formed through ‘NC-to-LNAT’ and ‘LNAT-to-Ice’ pathways respectively. It indicates that LNAT nucleated on the stratospheric aerosols, which subsequently acted as nuclei for ice formation. Moreover, the ‘LNAT-to-Ice’ conversion should have occurred rapidly to explain the high anomalous ice areal coverage. In fact, both ‘NC-to-LNAT’ and ‘LNAT-to-Ice’ pathways are parts of the three-stage PSC formation model (Peter, 1997). …

So what is new here? If we get past the LNAT acronym to remember what Pitts et al.’s classification is, we find it is a liquid-NAT mixture. The only liquid PSC is also known as STS so it is an STS-NAT or a sulfuric acid water and NAT mixture with a depolarization ratio too high to be STS alone. So as the air cools some NAT forms and at about 3K below Tnat STS forms giving us this mixture. Then if temperature continues to cool to about 6 K below Tnat some ice may form. The extent to which NAT and ice form is dependent on the availability of NAT and ice nuclei. These can be as the authors point out meteoric dust, sulfuric acid and water with unusual inclusions, or possibly hydrates of sulfuric acid, although these have never been observed in the atmosphere, or some unknown particle type. But all this follows the three stage PSC formation model as has been documented in the literature over many observational, laboratorial, and theoretical papers. What is new here? If PSCs form in any air mass they will be some combination of NAT and STS, e.g. LNAT in this paper’s acronym, and if it cools further ice is likely to form. How has this analysis shed new light on these processes? In fact the generalizations are so broad in the initial classification scheme that these statistical analyses of the temperatures and hno3 concentrations can hardly lead to any useful new information.

A Review of “Australian Bushfire Emissions Result in Enhanced Polar Stratospheric Ice Clouds” by S. Prasanth et al.

< General Comments >

This paper describes the impact of Australian extreme bushfire event during 2019/2020 on Antarctic PSC occurrence and subsequent ozone hole in 2020.  It is a new idea to investigate the influence of extreme bushfire on Antarctic PSC formation. However, since the author discuss only one winter (2020), the discussion and conclusions that the authors mentioned in the paper is not very convincing. At least, the authors should analyze another year when the Antarctic ozone depletion is about the same magnitude as 2020 (for example, 2021 or 2022), and compare the results with those in 2020.

< Specific Comments >

1) P.3, L.72: MLS is a 5-band microwave radiometer, not a Fourier-Transform Spectrometer (FTS).

2) P.3, L.83: The authors used the terminology “Liquid Nitric Acid Trihydrate (LNAT)” throughout the paper. However, Nitric Acid Trihydrate is a solid PSC, not liquid at all. I assume that the authors are referring “Mix 1” or “Mix 2” PSCs in Pitts et al. (2009), which are the mixture of STS and NAT PSCs. The authors should not use the term “LNAT”, but better use the term “Mix PSCs”.

3) P.7, L.204, Figure 3 caption: What is the definition of “standardized anomaly (Z)”? Please define Z by a equation.

4) P.8, L.214: What is the meaning of “additional”? In addition to what? Please explain.

5) P.8, L.222: What is the meaning of “additional”? In addition to what? Please explain.

6) P.8, L.222; Fig S1: The year 2019 was a exceptional year when the Antarctic ozone loss was minimum (like the year 2002), due to the dynamical effect (split of polar vortex in early spring). The authors should not compare the year 2020 with 2019, but had better compare with another normal year (like 2021 or 2022). One suggestion is to show time-series plots not for 2019-2020, but show like 2020-2021.

7) P.11, L. 292: The authors claim that “increased H₂O is transport-related as the data are close to the linear fit”. However, I felt that the 2020 H₂O data are also deviated upward from the linear fit.

8) P.13, L. 331: The authors claim that “Followed by ice, the Supercooled Ternary Solution (STS) exhibited a high positive anomaly”. However the presence altitudes of ice and STS are different (ice: 15-20 km, STS: below 15 km). STS are not “followed” by ice.

9) P.13, L.334: The authors claim that “the STS areal coverage can … lead to additional ozone loss.” However, since the appearance altitude of STS are mostly below 15 km, the additional ozone loss cannot be expected in such low altitudes.

10) P.14, Figure 9: I am curious if the authors also show the similar plot to indicate the formation pathways of STS from other types of PSCs.

11) P.17, L.422: In the course of the formation of ice PSC, why uptake of HNO₃ occurs? Where the decreased HNO₃ goes to? Please explain.

12) P.18, L.457: The authors claim that “explain the high anomalous ice areal coverage.” However, not reference nor supporting material/figure are shown to support the “high anomalous ice areal coverage” in 2020. Please show anything to explain that ice areal coverage in 2020 was anomalously high compare with other years.

13) P.19, L. 468: The authors claim that “anomalously high PSC areal coverage was observed.” However, not reference nor supporting material/figure are shown to support the “anomalously high PSC areal coverage” in 2020. Please show anything to explain that PSC areal coverage in 2020 was anomalously high compare with other years.

< Grammar/Typos >

14) P.2, L.36: Last “;” after Selitto et al. is not needed.

15) P.4, L.120: orbit the Earth ~15 times ---> orbit the Earth ~14 times

16) P.6, L.185: ΔT = T_n – T_n+2 ---> ΔT = T_n+2 – T_n

17) P.6, L.186: ΔHNO₃ = HNO_3n – HNO_3n+2 ---> ΔHNO₃ = HNO_3n+2 – HNO_3n

18) P.6, L.188: obtained from MLS and MERRA-2 ---> obtained from MERRA-2 and MLS

19) P.9, L.240: in k_ext ---> in Δk_ext

20) P.9, L.241: the result of mesospheric air ---> the result of the descent of the mesospheric air

21) P.10, L.276: (such as convection or advection. ---> (such as convection or advection).

22) P.10, L.281: chemical production of ---> chemical production or destruction of

23) P.11, L. 298: the same period (Fig. 7a) ---> the same period (Fig. 7c)

24) P.13, L.328: mid-April itself (Fig. 7c) ---> mid-April itself (Fig. 7b)

25) P.15, L.375: ΔHNO₃ = -0.8 ppbv --->  ΔHNO₃ = -1.0 ppbv

26) P.16, L.399: ΔT = 0.8 K --->  ΔT = 1.0 K

27) P.18, L.439: ΔT and ΔHNO₃ of -0.7 K, and -0.9 ppbv --->  ΔT and ΔHNO₃ of -1.1 K and -1.1 ppbv

The paper by Srinivasan Prasanth and colleagues addresses a very interesting and hot topic: The impact of bushfire events on Antarctic PSC formation, specifically the impact of the extreme Australian bushfire event in 2019/2020 on the occurrence of Antarctic PSC in 2020. The study "aims to investigate the anomalies in stratospheric chemistry and PSC dynamics caused by the Black Summer event" and tries to retrieve and quantify PSC formation pathways.

After reading the manuscript, I came to the conclusion that this study cannot be published. Therefore, my review focuses only on my main concerns along with a few examples.

I have the impression that the authors are newcomers to PSC research. A comprehensive understanding of the cloud formation processes seems to be missing. The data analysis is much too superficial. Averages are given for the whole year, making it difficult to justify the conclusions drawn. I would recommend to take this great data set of CALIOP, MLS and OMPS measurements together with ERA5 data and the CLaMS model and to improve the analysis after an intensive study of the literature.

Here are a few selected comments.

• Authors should adhere to established PSC terminology. NAT particles are solid and introducing "liquid nitric acid trihydrate (LNAT)" is misleading and won't be accepted by potentially interested readers of this paper. LNAT means that solid NAT particles are mixed with liquid STS droplets.
• The authors mix the different classification schemes of Pitts et al. The authors use the PSC product v2 (Pitts et al., 2018), but partially describe the classification from the PSC product v1 (Pitts et al., 2009 and 2011).
• Figure 1: I cannot believe that the scan at 13 UTC on August 01, 2020 shows no PSCs. It is the Antarctic winter. Why should it be a completely different picture 4 hours later? It is also difficult to reproduce the scene because the authors plot the observations with distance on the x-axis. Distance to the intersection? I could not find the LAT/LON coordinates of the intersection point, only the plotted orbits in Figure 1. Figure 2 must be a very small section of the orbits. If these intersections are the basis for the entire PSC formation analysis, I would like to see more examples, broader CALIOP orbits scenes and trajectories connecting the observations.
• It is impossible to infer anything about PSC formation pathways without looking at individual air parcels and trajectories. The authors use the CLaMS model driven by ERA5 data. This tool provides a high resolution picture of the polar vortex. Instead, the authors use the MERRA temperatures provided with the PSC data, but only at the point of observation. However, it is also necessary to look at the temperature histories along the trajectories between the individual observation points. This is really a key point that I want to emphasize! The authors cannot say anything about PSC formation processes without doing a trajectory analysis.
• The formation mechanism of NC -> PSC does not make sense. STS droplets do not form suddenly, they gradually increase in size from the stratospheric sulfuric acid aerosols by taking up HNO3 and H2O and at a certain size they can be detected by CALIOP. Most of the time NAT and ice particles are mixed with STS droplets. Especially when looking at "LNAT". How can LNAT be formed from NC without STS, since LNAT contains STS, otherwise it would not be LNAT?
• Another example is here: "If the same air parcel became populated with 'LNAT' after a certain time, this could imply that it formed either by nucleation on stratospheric aerosols favored by the decreased temperature, or by evaporation of large NAT rocks favored by the increased temperature, so that their size now falls within the CALIPSO detection thresholds". NAT rocks may not be detected by CALIOP because of their low number densities. Not because they are too big. As the temperature rises, NAT rocks evaporate and become smaller, but this does not change the number density. They won't be detected just because they're smaller. That makes no sense.
• “Furthermore, each formation pathway occurs at a specific temperature, which is conventionally viewed in the "T-Tice" temperature coordinate”. This is also far too simplistic. To give just one example, the cooling rate also has an important influence on PSC formation. If the temperature decreases slowly, PSC particles have time to grow. If the temperature decreases rapidly, many more PSC particles can nucleate but remain small. The result and also the PSC class will be different even if the observation point has the same temperature.
• How can the authors conclude that "Most of the LNAT (~82%) was formed by heterogeneous nucleation on wildfire aerosols"? They may be able to conclude that NAT formed via a heterogeneous nucleation pathway. But how do they know that these nuclei all came from the wildfire? Heterogeneous nucleation also occurs in other winters on other foreign nuclei of speculative origin, still a matter of research.