Review of "The impact of dehydration and extremely low HCl values in the Antarctic stratospheric vortex in mid-winter on ozone loss in spring" by Zhang-Liu et al., submitted to ACP

Summary: The paper addresses three questions: What is the impact of updates to previous recommendations on chemical kinetics on Antarctic ozone depletion? Furthermore, while dehydration strongly regulates Antarctic stratospheric water vapour, its impact on ozone depletion is small. And thirdly, an HCl null cycle and a further cycle starting with CH2O + Cl -> HCl + CHO contribute substantially to keeping HCl low and ClOx high, hence leading to enhanced ozone depletion.

I learnt a few things reading the paper. I had not thought about the two null cycles and their role in sustaining ozone depletion. The prevailing view is that CH4 + Cl is a termination reaction for ozone depletion, not the start of yet another cycle of ozone depletion and a null cycle for HCl.  Also the typo / order-of-magnitude error in the reaction ClO + CH3O3 is good to know about -- that might be wrong in many chemistry models. The paper represents good, solid work, enhancing our understanding of chemical kinetics of the Antarctic polar vortex. Of course this topic is sometimes considered to be fairly mature, but this paper presents a fresh take on this subject. I don't have many comments to make; the method is fairly straightforward. It involves trajectory calculations simulating atmospheric chemistry under Antarctic conditions and testing the sensitivity of the results to assumptions on initial values for HCl and water, and for correcting the typo in the rate coefficient. 

I recommend publication of the paper in ACP subject to addressing the small, technical comments below. 

Details: 

L17: You want to add that the temperature range refers to potential temperature, the vertical coordinate in CLaMS.

L23: Replace "although" with "notwithstanding".

L60: Conventional wisdom has it that NAT is important here too. Please comment. I suggest to replace "ice particles" with "PSC particles".

L116: Replace "on" with "to".

Table 1: Here and throughout the text, I suggest to put "volume" in front of "mixing ratio", and to use units of ppmv, ppbv, etc, instead of ppm and ppb. Otherwise these can be misunderstood.

Section 3.2 Can a line be drawn from the small impact of the initial value of H2O on chlorine and ozone to the (thus far) small impact of the increased water vapour in the stratosphere since the Hunga-Tonga Hunga-Haapai eruption? There had been some expectation in the community that this would increase ozone depletion, but the 2023 season was quite ordinary.

Review of Zhang-Liu et al. , The impact of dehydration and extremely low HCl values in the Antarctic stratospheric vortex in mid-winter on ozone loss in spring

General comments

The manuscript describes new modelling simulations of Antarctic ozone depletion using the well-regarded CLaMS model with meteorological fields from ECMWF.

Building on previous work by much the same team, the same techniques as previously used are again made use of to study the effects of making specific improvements to a number of the parameters of the simulation, namely updated reaction rates and more realistic values for water vapour and HCl as seen in observations. The authors find that with these changes, the model still simulates extremely low ozone in late September, as required to match observations.

While this could be seen as a null result which doesn't add very much to our understanding of polar ozone depletion, it is good science to investigate the effect of all such potential issues in previous work and to assess the sensitivities of the earlier results.

The subject matter of polar ozone depletion is central to the scope of ACP and I believe the manuscript is suitable for publication after some fairly easy revisions.

My only major concern is that, while I find the manuscript is very clearly written, in the sense that each individual sentence is well-written and easily understood, the broader narrative is not very clearly expressed.

I would like to see several points being better discussed for the benefit of the reader.

The abstract and the introduction need to explain better that this work is building on previous results. Similarly, the core method of using a single reference trajectory to evaluate the model output needs to be discussed (the readers shouldn't have to refer to the older papers) and the strengths and weaknesses of this approach outlined. The authors don't explain why unrealistic choices of water vapour (in particular) and HCl were used in the older work – according to section 2.2.1, two 2018 papers used 4.1 ppm but the observations listed giving a lower concentration were generally well known before 2018.

The authors also don't suggest any other ways the new simulations could be tested other than the effect on the ozone centration at the end of the reference trajectory – for example, wouldn't there be some observational consequences of the much greater surface area of ice clouds shown in figure 3?

Section 3.4 was not very clear in terms of the motivation, the details of the method or what exactly the results were showing, and needs some particular extra work.

There is some repetition in the text which could be cleaned up.

The references are very thorough.

Specific comments

Lines 2-11 Please re-write the abstract to better describe the purpose of the paper

Line 52 Please state the reactions you are referring to here

Line 55 Please state the phase of this reaction

Lines 56-57 please state or refer to the specific reactions

Line 75-76 "the uptake of HNO3 on ice particles" has not been mentioned until now

Line 87 This reaction has not been previously mentioned either

Lines 98-101 How does the model know what the surface areas are of these different types of clouds though?

Table 1 – how have these values been determined though? In particular, how have you decided the Bromine concentration?

Lines 120-139 The reader is left to puzzle why the 2018 papers used 4.1 ppm when there was such an abundance of observational data available to support a lower value – this point should be briefly discussed, otherwise it sounds strange.

Line 148 – I don't think you quite mean "it must be a process missing in the models". You next state it could be a temperature bias in the meteorological fields, which isn't a missing process in the models.

Line 220  What do you mean by "the occurrence heterogeneous reactions"?

Figure 2 – A general question about the method – the trajectory is calculated for months but its path could not possibly be accurately determined for such a long period of time – does this matter?

Lines 25-271 Section 3.4 is not explained well enough, you need to better motivate this section for the reader, explain what exactly the different trajectories are, and discuss what it shows.

Lines 296-298 This sentence reads very awkwardly at the moment and needs some minor re-wording. " … while ozone depletion is somewhat enhanced …  ozone depletion is not strongly affected"

Review of “The impact of dehydration and extremely low HCl values in the Antarctic stratospheric vortex in mid-winter on ozone loss in spring”

General comments

The paper entitled: “The impact of dehydration and extremely low HCl values in the Antarctic stratospheric vortex in mid-winter on ozone loss in spring” explores in depth the role of HCL null cycles on maintaining active chlorine in early Antarctic spring. The paper looks into the roles of initial wintertime HCL concentrations (where there is a known discrepancy between models and observations), a correction to the CLO + CH3O2 reaction rate, and dehydration on the HCL null cycles.

Overall, the paper is well written and is a nice addition to literature on Antarctic chlorine partitioning. Further knowledge on the HCL null cycles and the factors that affect them is an important and welcome advancement to the knowledge of Antarctic chlorine partitioning and ozone loss. I have a few comments below that I would like to see addressed.  I suggest publication after the following minor revisions.

Main comments

The authors present the majority of the results in a concise way, however I found the discussion around the CLO+CH3O2 reaction rates, specifically discussion of results of the incorrect A-factor analysis in the methods section, hard to follow. I feel this section can be shortened somewhat or made more concise, especially as results are discussed here but not shown (apart from a few values printed in text). The authors also state in the abstract that there is little difference between the two simulations when using the old (Sander) rates and new (Burkholder) rates. Looking at Figure 2 it looks to me that the differences can be quite significant between the two simulations and remains through to December 1. This may seem insignificant, but such differences after only one reaction is notable. This conclusion is a theme in the other cases investigated as well.

Is there no role of CLONO2+HCL in spring in maintaining elevated active chlorine? Your box model clearly shows no CLONO2 at all through to December. However, I believe there should be some elevated CLONO2 when spring arrives and therefore this reaction should also play some role. For example Solomon et al. (2015) Figure 3 shows elevated springtime CLONO2 levels from MIPAS observations. The reaction is likely not proceeding as fast as HOCL+HCL, but will the addition of this reaction affect the null cycles in any way? I feel this needs to be at least addressed in the paper.

Specific comments

Lines 53-55. Does CLONO2+HCL also play a role in maintaining elevated chlorine? 

Lines 179 and 182. I believe these equations should be A*exp(-E/RT) not A*exp(-R/ET)? Based on Figure 1 and Table 3 it looks like this is just a typo, but please check.

Section 3.1. I would like to see this section expanded on a little to explain why there are differences when changing from the older to newer rate recommendations.

Line 225. The authors state: “Further, a substantial difference in initial water vapour mixing ratios does not result in a substantial difference of polar chlorine chemistry and ozone loss (Fig. 4). There is a slightly lower minimum value of ozone (≈ 10 ppb lower) for an initial water vapour mixing ratio of 4.11 ppm.” Again this seems a quite significant change to me. Some discussion of why this isn’t would be welcome here.

Line 240-245. The earlier onset of ozone loss here is interesting and I would like to see it discussed more. This to me is quite substantial especially when early winter HCL conditions is something that fully coupled models can’t simulate accurately at the moment, as you mention in the paper.

Technical corrections

Line 193. Please remove “a” from “to a much larger value”.

Line 295. Suggest rewording “Further, while ozone depletion is somewhat enhanced under these conditions, ozone depletion is not strongly affected” as it currently sounds contradictory.

References

Solomon, S., Kinnison, D., Bandoro, J., & Garcia, R. (2015). Simulation of polar ozone depletion: An update. Journal of Geophysical Research, 120 (15), 7958–7974. https://doi.org/10.1002/2015JD023365