In this manuscript, the authors describe a set of multiple dynamical parameters for diagnostic studies of the upper troposphere – lower stratosphere (UTLS). This region is characterized by strong gradients in e.g. ozone, water vapor, or temperature in both the vertical and horizontal directions, and the use of the proposed dynamical parameters allows to better separate tropospheric and stratospheric air masses as well as air masses from either side of the subtropical or subvortex jets. This distinction help reduce uncertainties in monthly averages or in the comparison of measurements with different sampling patterns. Trends in tropopause altitude or subtropical jet position can also influence atmospheric parameter trends in this region. While the use of these dynamical parameters is not new, the manuscript provides an integrated view of their use and a detailed description of their calculation. The article is well written and documented. I thus recommend publication provided that the following comments are taken into account in a revised version.

Main comments

1. The benefit of using the diagnostic framework proposed in the manuscript for e.g. measurement comparison or trend evaluation is not clearly highlighted in the article. Figures 7 and 8 show some climatology comparison for ozone and temperature in the various coordinate systems but the advantage of using them is mainly based on citations (e.g. in page 13) while some demonstrations could be made in the article itself by showing for instance improved comparison time series or reduced error bars in monthly averages after using alternative coordinate systems.
2. The authors employ their diagnostic framework to some measurement time series e.g. a few ground-based sondes and lidar records, the MLS, SAGE III ISS and CARIBIC-2 time series and some campaigns measurements. It is not clear why these measurements were selected. In addition, some of the records are mentioned (e.g. SPURT, POLSTRACC, TACTS, WISE, START02 and most sondes and lidar records) but are not used in the article.
3. Some figures, e.g. Figures 4 and 5, are rather small and it is difficult for the reader to check the coherence of the patterns for the various records (Figure 5). In Figure 4, the distance from ground location for the satellite measurements needs to be explained.
4. Only 4 ground-based records are highlighted in Figures 7, 8, A and A2, while 14 ground-based records were processed for the article as mentioned in Table 1. The records’ selection is not explained in the article. It would have been interesting to show more examples of ground-based climatological data in different coordinate systems. I also suggest to include a table providing the equivalent latitude range covered by the various ground-based records or the latitude range with respect to the subtropical jet.
5. Differences in the climatological data displayed in Figures 7 and 8 need to be better described and explained. For instance, are the differences seen in MLS and SAGE III/ISS Theta/Eq latitude climatological data for equivalent latitudes > 60°S due to differences in sampling? In both figures, climatological data from CARIBIC-2 seem rather different. What is the reason for this and also for the differences in MLS and SAGEIII/ISS climatological temperature data?
6. In Figures 7 and 8, red and orange lines are difficult to distinguish. The black contours showing the wind speed are also different from one panel to the next. What are the contour values and what is the reason for the differences?
7. It would be useful to add some consideration in the conclusion about spurious trends in the reanalyses used to derive the diagnostic parameters and to which extent they could affect trend evaluation in alternative coordinate systems.
8. A lot of self-citations. The authors could reduce the number of self-citations, in particular the Manney et al. citations (15 in the current version).

Specific comments

P2 L30-31: what is the reason for the trends in jet altitudes and velocities?

P3 L6: to which measurement records did the satellite show a 10% bias in the zonal mean representation of ozone in the UTLS?

P3 L17: equivalent latitude is not a tropopause related coordinate.

P4 L30: results from JOSIE experiments could also be cited here (e.g. Smit, H. G. J., et al. (2007), Assessment of the performance of ECC-ozonesondes under quasi-flight conditions in the environmental simulation chamber: Insights from the Juelich Ozone Sonde Intercomparison Experiment (JOSIE), J. Geophys. Res., 112, D19306, doi:10.1029/2006JD007308):

P4 L34: what is meant by homogenized?

P5 L10: typo “typically”

P5 L16: some studies are lacking in the list, e.g. Hubert et al., 2016, cited somewhere else in the article. Also, references are mostly related to both TMF lidar systems and could be diversified.

P5 L25-35: The paragraph is difficult to read due to the number of acronyms. I suggest to refer to Table 2 for the campaign data.

P6 L22: explain better the link between “oversampling” and the nominal vertical resolution of 1 km.

P7 L10: the whole NDACC ozone sondes records could be cited here.

P7 L17: does the timeline of dynamical diagnostics correspond to the time range of the measurement records?

P8 L30-32: the sentence is not clear: which other code could be used to compute the diagnostics?

P9 L14: potential temperature has not been introduced before log(theta) is mentioned here.

P9 L22-23: the sentence is not clear, please reformulate.

P9 L28: MERRA-2 fields overestimate ozone and wind speed more than just slightly.

P9 L30: for non-specialists, please explain tropopause inversions, especially with respect to double tropopauses.

P10 L19: I thought that WMO tropopause is defined for a lapse rate below 2K km^-1.

P10 L28: please provide pressure levels for the thermal and 4.5 PVU tropopauses.

P12 L5: please explain tropopause break.  

P14 L12: it is not clear which fields the end of the sentence starting with “those fields” refers to.

P34 Table1: there are also two different lidar systems at OHP. Timespan for the stratospheric lidar is 1985 – present and that of the tropospheric is 1991 – present. The Hohenpeissenberg lidar records starts in 1987 and not 1978.

The paper is dedicated to several diagnostics for the UTLS studies, which are derived with the JETPAC software and MERRA2 reanalysis data. The authors introduce the SPARC OCTAV-UTLS project, which aims to reduce uncertainties in the UTLS trend estimates by accounting for dynamically induced variability. The authors state that “suitable coordinates should increase the homogeneity of the air masses analyzed together, thus reducing the uncertainty caused by spatio-temporal sampling biases in the quantification of UTLS composition trends”.

The paper has a comprehensive introduction and describes well the objectives of the OCTAV-UTLS project. However, the scientific information content of the paper is low. The paper describes in detail the computing of dynamical diagnostics using the well-established JETPAC software, which is already described in several papers.  Although examples of climatological ozone distributions in different coordinate systems are provided, there is no demonstration of advantages of using the alternative coordinates (references to published papers are not sufficient). In the current form, the paper does not meet the scope of AMT: “the development, intercomparison, and validation of measurement instruments and techniques of data processing and information retrieval for gases, aerosols, and clouds”.  The information related to atmospheric measurements, which is presented in the paper, is insufficient for AMT.

From my point of view, there are two ways of improving the paper. The first one (preferable) is to add more illustrations and discussion on datasets in various coordinate systems and their agreement.

Another way is shortening the paper and submitting it and the dataset of dynamical diagnostics to ESSD.

OTHER COMMENTS

• The link to the dynamic diagnostics data does not work.
• It is mentioned that the collection of the datasets used in the paper is limited. With the selected subset of available data, it is probably possible to study some processes in the UTLS. However, for the ambitious objectives of OCTAV-UTLS, the dynamical diagnostics should be provided also for other available satellite and in-situ measurements. Alternatively, JETPAC can be made as a free software.
• P.7 ” When computing dynamical diagnostics as discussed in this section it is important to use the same reanalysis fields for all the datasets to be used in a given study.”  However, possible discontinuities in reanalyses data, which are caused by changes in assimilated datasets (your discussion on Page 8), will also affect the dataset of dynamical diagnostics. When the same reanalysis is used for all datasets, these discontinuities will appear as an artificial drift in the dynamical parameters and thus in evaluated trends (if these dynamical parameters are used  for data  transformation). When using different reanalyses, one may hope that the timing of discontinuities will be different, and thus the overall evaluation of the trends using multiple datasets will have a reduced reanalyses-related drift.  These issues should be discussed/mentioned in the paper.

• The influence of sampling patterns should be discussed in more detail. In particular, it should be mentioned /discussed also in the text related to Figure 5 and Figure 7. The distributions for CARIBIC-2 data are different from those for other datasets.