This paper presents measurements from the recent DCOTTS aircraft campaign over the US, which aimed at sampling deep convective storms and investigating their impact on the UTLS. The focus is on one particular flight (31 May 2022) and on the convective enhancement of several tropospheric tracers and water vapor. It is shown that within the convectively influenced air the water vapor mixing ratios are significantly enhanced, up to about 2km above the local tropopause. Special emphasis is laid on the measurements within above anvil cirrus plumes (AACP), which were observed during that flight. Water vapor versus ozone correlations together with ice enhancements show that in the AACP air masses the convective moistening is saturation-limited, and that vertical mixing with the ambient air frequently occurs. 

I find this an interesting study on an extreme case of convective impact on the UTLS region which, together with the presentation of new observations, clearly falls within the scope of the journal. The paper is well written and the presentation is clear. I have a few minor and specific comments below which will hopefully further improve the paper and which needs to be addressed before publication.

Minor comments:

1. Correlation analysis (L350ff and Fig. 8):

I'm not entirely convinced how meaningful the relation between the calculated correlation coefficient and the dehydration/moistening process is. For instance, how to explain the frequent changes of correlation characteristics over short time periods? If a moist plume is characterized by both positive and negative correlations (e.g. the discussed about 23ppmv enhancement in the ascending leg of the profile around 02:25 UTC, or in the second AACP leg around 02:32 UTC) how to interprete that - or, in other words, by which process is it caused? Also, some of the positive correlations occur during flight times when the air is largely subsaturated (e.g. at the end of the first AACP leg around 02:22 UTC or during the descending profile leg) - so how can the related moistening process be interpreted as being "saturation-limited"? Could it be that additional noise in the timeseries blurs the physical signal in the correlation to some degree? Or could these issues be related to the "subjectively identified convection" and could be resolved with reconsidering the used criterion (see my specific comment below)? It would be good to add some sensitivity analysis  to show that  the calculated correlation coefficients are indeed physically meaningful, and/or to discuss the uncertainties and limitations.

2. Global impact (abstract, conclusions):

The relevance of convection and in particular AACP-producing storms for the global scale is to some degree discussed in the conclusions section. However, I find that this discussion could be enhanced. In this context, also the sentence in the abstract "Such UTLS composition changes alter the radiation budget ..." appears to me as if the impact of these convective events on global scale would be given. My personal impression here is that from a global perspective the occurrence frequency of these types of deep storms is not very high and the amount of tracers and water vapor which could be released into the stratosphere will be rather limited. Sure, a thorough investigation of this impact is an independent project. But are there any published studies which aim at upscaling the impact of such storms to global scale, and what impact could be expected? At least, the potential impact on the global scale could be discussed more critically here.

Specific comments:

L36: I'm somewhat surprised about this high fraction of overshooting convection reaching the overworld (50%). Can you add an appropriate reference for that?

L172: I'm also surprised that using 37 pressure level data instead of the original 137 level data has no effect on tropopause detection and the presented results. Could you explain further why, and could you state the pressure levels among the 37 which are in the UTLS region? Given the importance of the estimated tropopause altitude for the presented results it could be worth explicitly showing this insensitivity (e.g. in an appendix figure).

L226: What does "subjectively-identified convectively enhanced" here means? On what criteria was this identification based? Please clarify and describe in the manuscript. Could this subjective identification cause issues for the correlation analysis (see my minor comment above)?

L283: Please explain here in more detail how the "hydraulic jump" is characterized.

L299: It would be helpful to explain here more clearly what exactly is meant by "change in the physical/dynamic process"? I guess you refer here already to the onset of turbulent mixing, as discussed further below - but this could be mentioned here already.

L396: When looking at Fig. 7, the 23 ppmv moist plume seems to be entirely within the ascending leg of the profile. Please clarify the text here.

Review of Gordon et al., Detailed Examination of Upper Troposphere Lower Stratosphere Composition Change from DCOTSS Airborne Observations of Active Convection on 31 May 2022

General comments

The submitted manuscript presents and analyses detailed observations obtained from one particular flight of an aircraft campaign, during which, quite remarkably, measurements of atmospheric composition were made along a novel bisect of an above-anvil cirrus plume located well above the tropopause. This detailed study clearly adds to our knowledge of UTLS processes and composition and transport across the tropopause.  

The manuscript is very carefully and clearly written, and firmly within the scope of ACP and in my opinion is very suitable for publication. I have only quite minor comments.

It is not always easy to present the wealth of data collected during a well-instrumented flight such as this in a way that is clear and insightful to the reader, but the authors have tried very hard, in figures such as 4 ,6 and 7 with their choice of plots and helpful annotations. (Some reviewers might object to figure 8 as being too pedagogical but personally I liked it.)

The reasoning seemed logical in all places and based on the data collected, although of course, as discussed by the authors in the conclusion, it is not clear how far the findings based on one particular event can be extrapolated or be considered representative of the general process. Hopefully, further campaigns will be undertaken to see how reproducible the features found here are at different times and places.   

Specific comments

Although the introduction was nicely done, I felt the range of references was perhaps slightly narrow with too much of a focus on the research of the co-authors Some examples of recent work being carried out by other groups include:

Nugent, J. M., & Bretherton, C. S. (2023). Tropical convection overshoots the cold point tropopause nearly as often over warm oceans as over land. GeophysicalResearch Letters, 50, e2023GL105083. https://doi.org/10.1029/2023GL105083

Spang, R., Müller, R., and Rap, A.: Radiative effect of thin cirrus clouds in the extratropical lowermost stratosphere and tropopause region, Atmos. Chem. Phys., 24, 1213–1230, https://doi.org/10.5194/acp-24-1213-2024, 2024.

Clapp, C. E., Smith, J. B., Bedka, K. M., and Anderson, J. G.: Distribution of cross-tropopause convection within the Asian monsoon region from May through October 2017, Atmos. Chem. Phys., 23, 3279–3298, https://doi.org/10.5194/acp-23-3279-2023, 2023.

Khaykin, S. M., Moyer, E., Krämer, M., Clouser, B., Bucci, S., Legras, B., Lykov, A., Afchine, A., Cairo, F., Formanyuk, I., Mitev, V., Matthey, R., Rolf, C., Singer, C. E., Spelten, N., Volkov, V., Yushkov, V., and Stroh, F.: Persistence of moist plumes from overshooting convection in the Asian monsoon anticyclone, Atmos. Chem. Phys., 22, 3169–3189, https://doi.org/10.5194/acp-22-3169-2022, 2022. 

(To be clear, I am only offering these as examples, by no means insisting that you cite them).

I had to re-read section 3.1 a number of times to form a picture of the flight path with respect to the features of the convection event. This was not helped by the fact that in figure 1, yellow and blue circles are used to mark features, but in figure 2a, points "A","B","C" and "D" are used, and then points "P1", "P2" and "P3" in figure 2b. There are only state borders marked to help relate one figure to another (I assume they are state borders, they might be something else). My suggestion would be to add a schematic diagram sketching the flight path in 3 dimensions compared to the location of the anvil and the cirrus cloud. I think something like this would be easy to do and would help the reader grasp the situation much more quickly. I found figure 5 very helpful to orient the discussion.

In one or two places the authors assume knowledge of US geography that an international audience might not possess - for example I had to google "Texas panhandle", and no latitudes and longitudes are given anywhere.

Line 27 I think this sentence should be re-worded, I think your meaning is that "overshooting convection" as discussed in the paper is often associated with "severe convection" in the meteorological sense and hence "severe weather", but it doesn't read that way.

Line 36-39 I assume this statement is based on Chang et al. 2023, in which case it needs some qualification, because I would say it only applies to the overshooting events that were strong enough to be detected by the criteria used.  Weaker ones wouldn't make it up as high.

Line 49 Define "storm-relative winds"

Line 50 Define (very briefly) "hydraulic jump"

Line 122 Please give the latitude and longitude for Salina.

Line 171 Is 250 m uncertainty good enough for the analysis that follows?

Line 187 You should give the central wavelength of these channels, not just "VIS" and "IR" and the channel numbers.

Figure 3 I think it would help to give the direction the camera is pointing here as well as in the text. (I have to admit I'm not really seeing much in these photos).

Lines 240-243 This isn't quite true for CH4 is it? It doesn't seem to drop between segments 1 and 2.

Figure 4 Ozone is included in the plot but not discussed in the text in this section (lines 240-275).

Figure 5 I found this figure very helpful.

Line 298 I don't understand what you mean by "a H2O dominant convective signature" – do you mean it looks like there has been convection of H2O but not other tropospheric tracers?

Line 325 Not "below a value of 1" but "well below", it's more like half.

Lines 370-376 I find your explanation plausible, but am not convinced that it would lead to a positive correlation.

Line 429 "desperately" seems a little bit overwrought.