Dear Dr. Gutjahr,
Your manuscript has been significantly improved. Both reviewers have some further suggestions. Please take them into account in your next revision, in particular the major comments from the first reviewer.
Best regards
Qiang Wang

------
Review of the revised manuscript "Comparison of ocean vertical mixing schemes in the Max Plank Institute Earth System Model (MPI-ESM1.2)"


I think the revised manuscript is greatly improved as compared to the previous version in terms of clarity. But I still have a few concerns.

I may misunderstand the main purpose of this manuscript. But based on the title and abstract I was expecting to see a more in depth discussion of the differences among the four vertical mixing schemes and how these differences lead to the differences in the simulated mean climate states in MPI-ESM1.2. However, it looks like that the authors would like to present the differences in the simulated mean climate states with the four vertical mixing schemes on a high level without going too deep into the causes of these differences. This is fair as I understand that it is hard to disentangle the climate responses to the changes in parameterizations in a coupled system. But I would suggest the authors to pay more attention on the differences between the four simulations with different ocean vertical mixing schemes, and perhaps a bit less on the common "biases" of these simulations. Otherwise I don't see much added value of this manuscript to the authors' previous publication Gutjahr et al. (2019), which, if I understand it correctly, is an overview of similar MPI-ESM1.2 simulations with a set of different configurations.

In particular I think there are a few pieces of information missing in the introduction section:
1) What is the main purpose of this study?
2) Why are these four schemes chosen, given the many other choices? Or why do you think a comparison among these particular four schemes in MPI-ESM1.2 is helpful?
3) What do you want to get out of this comparison of the four vertical mixing schemes in MPI-ESM1.2?
Then, in the presentation and discussions of the simulation results, I would suggest the authors to focus more on differences among these simulations. The present discussions appear a bit unfocused to me.


Specific comments:

L41-50: A sentence or two on why the four schemes were chosen given these many more may be helpful. Perhaps moving the paragraph of L29-35 here after this paragraph?

L64: "ist" -> "is"?

L69: "modified version PP" -> "modified version of PP"?

L82: What is the main difference from the CVMix version? Is the Marsland et al extension the only difference?

L110: "first" -> "upper"?

L115: Based on Table 1 it looks like you are using PP below the mixed layer for HR_kpp. Does that mean the only difference between HR_pp and HR_kpp is the treatment in the mixed layer? How is the mixing in the mixed layer treated in HR_pp?

L149-150: But it looks like the deep convection in the Labrador Sea is the strongest in HR_ide? What might be the reason?

L171: I can see quite some differences in, e.g., the Eastern and Western Equatorial Pacific and the subtropical regions in the Atlantic?

L182: The simulations are forced by constant 1950s conditions whereas the PIOMAS represent the average of 1979-2005. Do you expect they will match each other if the model is perfect?

L184-186: Are you comparing with HR_pp? HR_ide seems to have an ice edge to the further north.

L188: By TKE here are you referring to both HR_tke and HR_ide?

L204-205: Comparing which vertical mixing schemes?

L206: Fig 8 introduced before Fig 7?

L211-212: What about the salinity bias in the South Pacific?

L213: Is the streamfunction in Fig 7 integrated in the vertical over the full depth, or from the surface to some depth?

L219-220: You can probably see this by looking at vertical sections of the salinity in this region.

L229-231: This reflects one of the issues in this set of simulations. If you run the simulations (forced by constant 1950s conditions) for longer, you might see different patterns of the "biases", right? So how sensible is the magnitude of the "biases" shown here in these figures?

L233: Delete "of"?

L234-237: I’m confused, are you discussing Fig 7 in this paragraph?

L245-248: This paragraph is confusing.

L258-260: I don’t understand this sentence.

L361: Do you mean "warm halocline layer"?

L436: Delete "but"?

L438-439: What do you mean by "in addition, we compare them with control simulations"?

Fig 1: This figure shows the differences between the four vertical mixing schemes in the ocean interior very nicely. I wonder if it is worth to show their differences in the mixed layer as well, which may be more relevant to the ocean surface states such as SST, SSS and MLD.


References

Gutjahr, O., Putrasahan, D., Lohmann, K., Jungclaus, J. H., von Storch, J.-S., Brüggemann, N., Haak, H., and Stössel, A.: Max Planck Institute Earth System Model (MPI-ESM1.2) for the High-Resolution Model Intercomparison Project (HighResMIP), Geosci. Model Dev., 12, 3241–3281, https://doi.org/10.5194/gmd-12-3241-2019, 2019.

------

Review #2 of:" Comparison of ocean vertical mixing schemes in the Max Planck Institute Earth System Model (MPI-ESM1.2)." by Oliver Gutjahr et al.

The paper compares the effects of four different ocean vertical mixing schemes provided by the Community Vertical Mixing (CVMIX) library on the mean state simulated by the coupled model MPI-ESM1.2. Used are the Pacanowski and Philander (PP), K-profile (KPP), turbulent kinetic energy (TKE) vertical mixing schemes as well as a prognostic schemes for internal wave energy and its dissipation (IDEMIX) which is combined with TKE. The author addresses temperature and salinity biases in a global and regional context with sufficient analysis of there causes and impacts. Compared to the first revision round the paper has significantly improved in scientific impact but also in its phrasing and expression. All comments of the reviewer have been considered. The paper was restructured for a better understanding and new chapters for the impact of the mixing schemes on the sea ice but also on the coupled atmosphere have been added. These chapters are of particular value for coupled modeling community.

From my point their are only minor things to change (linenumber with respect to file gmd-2020-202-ATC1.pdf):

line 67: "...submodel JSBACH3.2, and of the ice-ocean...", delete of

line 73: "...River runoff ist calculated...", exchange to "is"

line 79: "...is a modified version PP scheme..." exchange with "...is a modified version of the PP scheme..."

line 110: "...wave-wave interaction is accounted for that might...", please reformulate

line 153: "...approximately homogeneously distributed to leading order...", not fully clear what the author want to say here

line 203: "...Ice thickness is lower in all simulations than in PIOMAS..." exchange with "...Ice thickness is lower in all simulations compared to PIOMAS..."

line 226: "...where they affect to local...", exchange with "...where they affect the local..."

line 278: "...across the Greenland-Scotland Ridge at and ...", delete at