This study investigates mountain waves in ICOsahedral Nonhydrostatic (ICON) model simulations, using two different turbulence schemes and various horizontal resolutions. Results are compared to DEEPWAVE campaign observations and to ICON-LES simulations.

The first turbulence scheme is the operational turbulent kinetic energy (TKE) scheme, which is comprised of several components, including optional horizontal shear (HS) and subgrid-scale orography (SSO) terms. The second scheme is the newly developed two-energy turbulence scheme (TE). Results of those simulations are partially compared to a flight performed during the DEEPWAVE campaign on the 12 of July 2014.

The manuscript is well written and includes many references. The researched topic is very current and important as high resolution simulations are increasingly common, and mountain waves play a crucial part in atmospheric dynamics. However, none of the two goals mentioned in the abstract, comparison of the different turbulence schemes and comparison to the flight measurements, seems to be sufficiently reached. For this reason, I recommend the manuscript for major revisions, with specific comments below.

Major comments:

1. Although the different simulations are very well formulated to achieve a good comparison between the two turbulence schemes and various resolutions, they are very limited by looking at a one specific timestep. This was chosen as to compare to the DEEPWAVE flight done at this specific time. However, looking at the one timestep is not enough to make conclusions about the differences between the two turbulence schemes, especially considering there are other times of flights available. If the primary focus of the paper is to compare the schemes and resolutions, I would recommend looking at more timesteps and at longer time averages, especially when it comes to variables such as momentum flux, that show significant differences between the various resolutions and schemes as well. This way the results would be way more robust.
2. There is a little comparison done between the observations and the simulations as such. The only figure that compares them is Fig. 4, which does not include the TKE simulations. Once again, to make the comparison more robust, it would be useful to include another flight, e.g. F10 (ref. Fritts et co. 2015), and another simulation time, or to include some quantitative measurements such as momentum flux calculated from the observations (ref. Smith and co. 2016).
3. It would be useful to include more theoretical and technical details regarding the different turbulence schemes and their components. This would help with the conclusions as they are not quite clear. For example, you show (Fig. 8, Fig. 9 and 10) that the I10-TKE simulation, which includes both HS and SSO components, has more similar wave field to I10-TKE-SSO than to I10-TKE-HS. However, the outputted SGS-TKE from the schemes is more similar between the I10-TKE and I10-TKE-HS[Neznámý a1]. The outputted TKE suggests that HS component of the scheme is more influential, but that is not consistent with the wave and wind field. Then, in Fig. 9 a) you can see that vertical velocity variance of the 0.5 km TE resolution simulation is closest to the LES simulation, but that’s not the case for the momentum flux in Fig 10 a), where those two simulations are the most dissimilar. Also, in the text on the line 287, you say that 0.5 km simulation converges to the LES one, which is not apparent from Fig. 10. More technical details regarding the schemes could help in explaining those differences, as they are not clearly explained now.

Minor comments

1. On line 169 you define overbars as horizontal mean, is it taken over the whole domain area?
2. Could you rephrase the sentence starting on the line 197? The meaning is not quite clear.
3. In the section 3.2 (line 215) you refer to Fig. 6 and after that you refer to Fig. 5 (line 228). Also, a paragraph starting on the line 235 refers to Fig. 4 and includes lot of information that was already said in a paragraph starting on line 201. This makes the text hard to follow in this section.

Technical comments

1. In the title of subsection 3.4.1 you introduce new acronym – SGS-TKE – which is not explained previously.
2. Figure 9 has incorrect description – subfigures c) and d) are not referred to and a), b) are referred to erroneously.
3. Choice of colours for the lines in all figures in Fig. 4, Fig. 9 and Fig. 10 makes them very hard to read and to differentiate between the simulations.

References

Smith, R. B., and Coauthors, 2016: Stratospheric Gravity Wave Fluxes and Scales during DEEPWAVE. J. Atmos. Sci., 73, 2851–2869, https://doi.org/10.1175/JAS-D-15-0324.1.

Fritts, David & Smith, Ronald & Taylor, Michael & Doyle, James & Eckermann, Stephen & Dörnbrack, Andreas & Rapp, Markus & Williams, Bifford & Pautet, P.-Dominique & Bossert, Katrina & Criddle, Neal & Reynolds, Carolyn & Reinecke, P. & Uddstrom, Michael & Revell, Michael & Turner, Richard & Kaifler, Bernd & Wagner, Johannes & Mixa, Tyler & Ma, Jun. (2015). The Deep Propagating Gravity Wave Experiment (DEEPWAVE): An Airborne and Ground-Based Exploration of Gravity Wave Propagation and Effects from Their Sources throughout the Lower and Middle Atmosphere. Bulletin of the American Meteorological Society. 97. 150709110621006. 10.1175/BAMS-D-14-00269.1.

This research study focuses on simulating mountain waves within the ICON model. The focus is on mountain-wave dynamics and turbulence in the numerical model and its dependence on turbulence paramterization and resolution. The operational turbulent kinetic energy scheme (TKE) and the new two-energy turbulence scheme (2TE) are applied in this study. For the first scheme, the impact of the individual components (horizontal shear (HO) and subgrid-scale orography (SSO)) are investigated. The outcomes are then compared against aircraft observations from the DEEPWAVE field campaign, as well as high-resolution ICON-LES model results.

Overall, the paper is clearly presented and well-supported by literature. The topic is highly relevant given the growing use of fine-scale atmospheric simulations also in the upper troposphere and lower stratosphere region. However, I have one main point - which is the lack of information and investigation considering the 2TE scheme - which is missing in the manuscript. Consequently, I suggest major revisions to address it.

Major comments:

• I am missing a discussion on why is the 2TE scheme is more appropriate. What physical aspects are responsible for this behaviour? You switched on and off the HS and SSO in the TKE scheme, can you do a similar analysis with the 2TE scheme?
• Further, no equation is offered here for the 2TE scheme (which should be included, at least in the Appendix). It is required to make some statement about what is it, which is leading to the better performance of the 2TE scheme.
• Concluding, after reading the paper I know why the TKE scheme fails going to fine resolution but I don't know what exactly is the difference in the parameterization of the 2TE scheme resulting in a better performance. This would be very interesting and might also be the main part to see what is missing in some models or how can they be improved.

Minor comments:

• Think about referring to Fig. 1 earlier in chapter 2.1.
• Line 215: At 2 km - where? The plot shows the vertical wind w - it is not clear why now a horizontal wave length is discussed.
• Think about adding Two Thumb Range and Mount Cook as names on the x-axis. It would help the reader.
• Fig. 7 and corresponding discussion:  It is not clear if it is the resolved TKE, the SGS TKE or the sum of both? ->SGS TKE comes in Chapter 3.4.1, but the first time mentioned here it is not clear.
• Fig. 8: Think about, maybe difference plots between the corresponding simulations in Fig 8 and the reference one are a way to be more quantitative?  - It would help chapter 3.3.2 and might be a way to include percental differences or something more quantitative in the text.
• Fig. 9: did the difference between I05-2TE and I01-LES-S matter in between 7-10km? Please try to be more quantitative instead of stating "in good agreement".
• Line 287: In the 2TE simulations, the 0.5 km run converges toward the LES, while the 2 km run shows reduced fluxes, especially in the lower troposphere. I am confused, is this sentence correct? Does it state that the dark blue curve converges towards the gray one? Aren't the differences in the lowest 4km important and should be interpreted and mentioned?

• 292: Overall, the results highlight that area-averaged momentum flux converges more rapidly with resolution than local turbulence diagnostics, making momentum flux a more robust diagnostic for evaluating mountain-wave simulations. Do you mean at lower heights? This is not clearly enough written for the reader to understand what you wanted to say and should be formulated in more detail and in a more quantitative way.

Technical corrections:

• 190: Further
• Fig 3: Think about making the dashed lines thicker. Especially the red ones overlying with the isentropes.
• observations not in the legend in Fig. 4
• reference to Fig. 6 before reference to Fig. 5
• 222: 100 km got some issue with a ~
• Fig. 5: labels and numbers are small especially in comparison to the other plots
• 250: Figs. 7a, c -> spacings are missed, also Fig. A2 in the same line
• 251 Figs. 7b, d