review of manuscript acp-2022-783

Title: Zugspitze ozone 1978–2020: The role of stratosphere-troposphere transport

Authors: Thomas Trickl et al.

General comments:

The manuscript presents an interesting and extensive analysis of more than 40 years of ozone data measured at the Zugspitze summit and the nearby Schneefernerhaus research station in the German Alps. The present manuscript builds on a series of publications by some of the authors. The manuscript reveals some interesting phenomena like that increasing stratospheric contribution to the O3 levels at Zugspitze over time. However, the paper is hard to follow as it is rather wordy, uses some vague terminology, mixes previously published results, methods and new findings, and includes an excessive number of figures. See some examples in the specific comments below.

To my mind, the paper is within the scope of ‘Atmospheric Chemistry and Physics’ but can profit from considerable shortening and removal or movement of some of the figures to the supplementary material and, thus, requires major revisions prior to its acceptance for ACP.

Specific comments:

Line 19: it is confusing that changes between 1970 and 2003 are reported while the title of manuscript refers to the 1978 to 2020 period.

Lie 21ff: "The air masses hitting the Zugspitze summit have become drier in the percentile range up to at least 25 % until 2003 …" difficult/impossible to understand. Please rephrase.

Lines 23-24: "… a growing width of the intrusion layer departing downward from just above the tropopause must be taken into consideration …" difficult to understand here (and on page 20)

Line 25: "… perhaps a slight increase …". Very vague wording. Is it statistically significant?

Lines 46 ff and Fig. 1, lower panel: It is uncommon to refer to figures already in the introduction. I do not see the added value of showing the Wank data here. It does not really contribute to the discussions and conclusions drawn here.

Lines 89-90: add reference to the "far below 1%"-statement.

Lines 121 ff.: Awkward sentence (in grammatical terms). Please rephrase.

Lines 179-179: add references or urls, according to the WCC webpage (at Empa), there was another audit in 2020.

Line 182: add url of the data centre.

Line 199: add location or url for Deuste-Steininger.

Line 201: add year 2020? See above.

Lines 217-2018: "The periods are time shifted to 7:30 to 19:00 CET and 19:30 to 7:00. This does not matter since the data filtering is carried out on a half-hour basis." Difficult to understand since the filtering is only explained below.

Line 220: "… both averages determined on a half-hour grid …" This is confusing. What does it mean? 7Be were measured as 12h or 24h bulk samples.

Lines 220-221: "This means that the 12-h DWD data were calibrated to match the 24-h IFU data. It could not be clarified why this is the case." What does that mean? The DWD were scaled to match the IFU data? You cannot call this a calibration. Your last sentence is weird … and I do not understand to what it refers. To the application of a scaling factor? If so, don't you think that it was done to homogenize the two records?

Line 228-229: Something is linguistically wrong in this sentence.

Lines 231-232: "… which turned out to be highly necessary near the temporal boundary (end) of our analysis." Weird wording. How about "… which were useful to apply to the filters (see below) to the whole dataset."

Chapter 3 is can be considerably shortened as it largely repeats previous findings. However, it also mixes methods, previous findings and new results/interpretation, which make it really hard to follow.

Line 284: "The specific activity per intrusion has changed over the years …" why?

Line 347: what is a short gap?

Lines 361-363: Move this statement somewhere upfront.

Lines 387-402: could be summarized by one or two sentences.

Lines 407-411: isn't it trivial when 7Be is steadily increasing as pointed out in line 378? Remove Fig. 4?

Line 435: " The monthly minimum RH …" This is the lowest half-hourly RH reading for each month, right? Is this a useful number? I don't think so. You even state that it might be subject to different sensors. Remove from Fig. 6.

Line 485: did Yuan et al. measure in a tunnel? Does that make sense?`

Line 718: Stohl, 2001: reference does not exist in the bibliography.

Lines 718-720: did you compare with CO data from IAGOS?

Line 750: Claude, 2003: reference does not exist in the bibliography.

Lines 756 ff: "… increasingly wider layers have separated from the range just above the tropopause …" Where is this conclusion coming from? Add reference.

Table 1: What does year "1970.50" etc. mean? Simply write "1970" …?

Table 1 and Figs. 14 and 15: "annual ozone averages in intrusions …". Maybe I missed it in the manuscript but these numbers have to be understood as "annual mean contributions to the ozone levels", correct?

Figure 12, caption: criterion needs to read "RH-criterion", not "7Be-RH-criterion", right?

Fig. 14: this figure (along with the figure caption) is very confusing when first looking at the figure without reading the whole manuscript as it is makes the reader think that O3 is depleted during stratospheric intrusions (since numbers here are lower than in e.g. in Fig. 1 top panel).

Figs. 16, 17, 18 and 20: reword captions to make it clear what is shown (O3 mole fractions at Zugspitze and their individual contributions).

Fig. 2: it is textbook knowledge that O3 and 7Be rise while rh drops during stratospheric intrusions. Does it really need a case study figure here?

Figs. 6 and 7: delete one of them or move at least to the sup mat.

Figs. 8 and 9: delete one of them or move at least to the sup mat.

Figs. 16 and 17 are pretty much redundant.

Review of Zugspitze ozone 1978–2020: The role of stratosphere-troposphere transport

By Thomas Trickl, Cédric Couret, Ludwig Ries, and Hannes Vogelmann

This presents an extensive analysis of ozone and related measurements at an important mountaintop site in Europe. The topic certainly appropriate for ACP. Unfortunately I think that the manuscript needs major revision before acceptance to ACP.

The principal issue is that in its current form the manuscript is much too long, contains too many figures (many whose value is not clear) and reads like a rambling conference presentation, with too much discussion of some aspects of the analysis while important details are omitted. Since the authors have published many excellent papers in the past, I have no doubt that a rewrite will fix these issues and present a tight, well-reasoned argument for their interesting conclusions.

Detailed remarks:

Line 30: The Introduction is rather long, and contains a lot of (albeit interesting) information that is not needed to set out the scientific question(s) to be addressed by the research.

Lines 64-74:  The 10 ppb value for the 19^th century has been discounted for many years by modelers, who point out that there is no way to have so little ozone in the troposphere unless trees weren’t producing isoprene in the 19^th century. The Tarasick et al. (2019) paper reviewed many other observations that together suggest a value of 25 ppb for rural (background) air. This paragraph, like many others, seems unnecessary to the point of the paper.

Lines 75-81:  More to the point, more STT is expected, because it is a function of the Brewer-Dobson circulation, which is expected to increase with climate change (Butchart et al., 2006; Butchart, 2014).

Lines 89-94:  Osman et al. (2016) attempted to address this question theoretically.

Line 214:  “standard deviation of the mean” is not a statistical measure. Do you mean standard error of the mean?

Lines 219-221:  “The IFU and DWD 7Be data agree rather well.” The DWD data are mentioned here without introduction. But apparently they were scaled to the IFU data, and “It could not be clarified why this is the case.” What is that supposed to mean? It does not inspire confidence in the results.

Lines 257-258:  “We prefer the “reanalysis” mode that have better explained our observations.” This is an odd scientific justification. Do you mean you used reanalysis meteorological fields rather than forecasts? The former would indeed be more accurate.

Line 265: This entire section reads like a rambling discussion of the authors’ thinking about different thresholds for data filtering, but does not offer any clear justifications for the numbers chosen. In the last paragraph (lines 324-327) the authors seem to be saying that the high 7Be numbers on July 27^th must indicate older stratospheric air, but the argument is not clear.

Line 344: Rather than this unscientific justification (the values “look somewhat low”), refer to the discussion of Fabian and Pruchniewicz’ data in Tarasick et al. (2019).

Lines 407-411:  I really can’t make head or tail of what Figure 4 is about. This entire discussion (lines 360-422) seems to be solely for the purpose of justifying using percentile changes in 7Be data, which are clearly a more robust indicator given the other variables involved. This section could be reduced to one or two sentences.

Equation 1 is given without derivation or explanation. The last term, a quantity to the 8^th power, seems absurd. This section (3.4) also seems to go into unnecessary detail about the variability of sensors and the like. RH data are typically of doubtful absolute accuracy, but like the 7Be data, should be useful as an indicator of changes.

Equation 2 looks fairly simple, but the explanation is quite complicated.

Lines 631-634: Too many significant figures! Your trends are not accurate to 0.001%.

Figures 11 & 12 show similar trends, but apparently little correlation. That suggests that the trend agreement is coincidental. Please put them on the same horizontal scale so the reader can tell if they do agree. Wait, you did that in Figure 14. Please DELETE Figures 11 & 12!

In summary, please reduce the length of the text and the number of figures considerably. There is an interesting message in this analysis but it is difficult to find.

References

Butchart, N., A.A. Scaife, M. Bourqui, J. de Grandpré, S.H.E. Hare, J. Kettleborough, U. Langematz, E. Manzini, F. Sassi, K. Shibata, D. Shindell, and M. Sigmond, 2006: Simulations of anthropogenic change in the strength of the Brewer-Dobson circulation. Clim. Dyn., 27, 727-741, doi:10.1007/s00382-006-0162-4.

Butchart, N. (2014), The Brewer-Dobson circulation, Rev. Geophys., 52, doi:10.1002/2013RG000448.

Osman, M.K., W. Hocking and D.W. Tarasick (2016), Parameterization of large-scale turbulent diffusion in the presence of both well-mixed and weakly mixed patchy layers, Journal of Atmospheric and Solar-Terrestrial Physics, 143-144 , 14-36, https://doi.org/10.1016/j.jastp.2016.02.025.