Please address the specific points raised by the referees on the revised manuscript.

#1

This is a review of a revised manuscript on the introduction and documentation of ESM-Tools by Dirk Barbi and co-authors. ESM-Tools aims to provide the German climate modelling community with a standard, easy-to-use/adapt ‘wrapper’ from downloading the model code all the way to model output post-processing. The tool is expected to ease the technical burden for climate researchers, especially for young researchers and those relatively new to climate modelling. As earth system models and their component coupling are getting more complex, and even formidable in some cases, I personally appreciate the initiative and the approach taken by the authors, and this manuscript, together with their website and online documentation, would serve as a very useful reference for the general climate modelling community, especially for those working with e-infrastructures.

The revised manuscript is overall in good shape and fits the scope and reader interests of GMD. However, in my view some revision/clarification is still needed before the manuscript can be accepted for publication. I have my specific comments listed below (most of them are minor though), followed by some technical corrections such as typo and grammar, etc.


Specific comments:

> Title:
Since version 5.0 of ESM-Tools has been released, I wonder if it would make sense for the authors to document the latest version, provided that the upgrade from 4.0 to 5.0 is not too dramatic, such that a significant part of the text needs to be updated/revised. This is up to the authors to decide.

> I’d suggest the authors state upfront that ESM-Tools is not a coupling software early in the Introduction and/or Abstract. I thought (or at least was in doubt) it has this capacity until I read later in the manuscript. Such a statement can be added at, for example P1L18 and P2L37.

> P1L11: …the ESM-Tools “are”…
The authors treat the ESM-Tools randomly either as singular or plural throughout the whole manuscript. Please make it consistent (I think it should be a singular noun).

> throughout the text, the authors use inconsistent references to the figures, e.g. fig x, fig (x), Figure x. Please make it consistent and comply with GMD format if any.

> P1L12-13: In fact, … standalone simulations.
I don’t see this sentence as important info in the abstract, as the previous sentence has mentioned this point. If the authors decide to have it, more information on FESM2, ECHAM, ICON is needed, for example at least on what components they are (ocean or atmosphere or). Full names are usually needed but that would make the abstract tedious.

> P2L29-30:
I don’t quite understand when the authors say that ‘ESMs are usually written for a specific purpose’. It reads ambiguously to me; do the authors mean ‘configured for a specific purpose’?

> P2L45:
I don’t understand what ‘script-based coupling’ is.

> P3L71:
Do the authors mean ‘nesting’ of model components here? Since it is a tool for coastal ocean modelling, it should only have the ocean model component, right?

> Figure 1:
I don’t see the point to have this as a figure. These are very basic lines and the description in the text is sufficient. In addition, the caption is a bit colloquial, e.g. it could be ‘Some basic requirements need to be met before…’

> P6L148:
I agree the tool would save lots of time for researchers, but I don’t see how the tool would have the ‘potential of reducing model diversity’.

> P6L154:
This sentence seems to be coming out of the blue, especially ‘solving a model run error’?

> P9L204-244:
This part of the text does not fit the scope of this subsection (2.2.2 YAML Configurations) and should be moved elsewhere, for example in the discussion.

> I also find the tool very useful for educational purposes and for provenance/reproducibility of model simulations. The authors have mentioned the latter, but perhaps that can be stressed even more.

> P9L215:
I don’t understand this ‘compile time switches’.

> Table 3 seems to be not referred to in the text at all.

> P12L270-271:
This is up to the authors to decide if they think it could be useful: it would be good to elaborate a bit more on the tool’s log file and post-processing capacity. For example, for the log files, does the tool collect all the log info from each component and put/condense them together, or most log info still stays with their respective component? For the post-processing, does the tool offer things like netcdf conversion, climatology/time series calculation, or even cmorisation to name a few? The feature of online automatic cmorisation would be very useful for many.

P12L275-276:
I like this esm_vis feature a lot, but I don’t agree on this point, since modellers can always sanity check the simulations anytime without having to wait for the whole run to be completed. I think the main advantage is that it is very convenient/handy to use and save modellers’ time.

> P19L387-398: these two paragraphs are almost identical to the first two paragraphs in this section? Please check.



Technical corrections:

> P1L20: move ‘lately’ up in the sentence, e.g. ‘Compared to previous versions, ESM-Tools has lately been …’

> P2L42: The abbreviation HPC should come earlier.

> P2L51: please give the full name of SMHI.

> P2L55: ‘This approach is also…’

> P3L67-68: should be updated; also see my first comment.

> P3L78: ‘in our opinion it lacks…’

> P3L84: remove ‘the’

> P4L89: please fix the ref for Righi et al.

> P4L101: ‘The tool’s application…’

> P4L105: ‘…applying an ESM…’; the full name has been defined at the beginning of the manuscript.

> P5L119, 122: ‘dependent’

> P5L127: remove ‘are’

> P5L137, and elsewhere in the manuscript: no space before and after ‘/’

> P5L138: ‘…, and some features will be…’

> P7L161: should the footnote to YAML come earlier in the draft where it first appeared?

> P7L163: ‘…, while the ESM-Tools…’ grammar issue; please consider starting a new sentence.

> P7L174: ‘organize’?

> Table 1, in the row of esm_calendar: ‘’but also works…’

> P8L178: ‘to alter’

> P8L180: ‘this dictionary’

> P8L181: ‘execution of a compute job’

> Figure 3 caption: ‘Section of…’

> P8L192: ‘to generate’

> P8L192: ‘and in this sense can enable…’

> P10L233: please try to avoid words like ‘nicely’

> P11L242: ‘for us’, the authors mean ‘core developers’? and the previous ‘developers’ are ‘contributing developers’?

> P12L279: please avoid using ‘you’, ‘your’. This also applies to P16L339 and P17L346.

> P12L280: ‘the remainder of the supercomputer’? not clear to me

> P14L307: ‘and the code’

> P16L316: ‘the tool, which…’

> P16L319: remove ‘even’?

> P16L342: ‘configuration’; ‘As an example, …’

> P17L347: ‘see an example again from…’

> P18L355: ‘… Project phase 6, CMIP6, …’

> P18L371: ‘, which were discussed…’

> P18L372: ‘point out that…’

> P18L378: ‘as if it were…’

> P19L383: ‘, as we provide…’

> P19L407: ‘webpage’

> P20L409: ‘upcoming’


#2

I am happy with the changes that the authors made; I think that the revised version of the manuscript is much improved. The level of detail provided has been risen, the structure is better now, which aids in the understanding, also due to additional text such as in Section 2.1. My remaining comments are minor and listed below. I also recommend to check grammar, punctuation, and consistency in detail. Apart from the remaining minor issues, I would recommend a publication of the manuscript.
With kind regards.

L16: Perhaps add something like "once implemented with ESM-Tools"
L34: "requires"
L98: Remove the bullet point, as this is not part of the listing.
L103f: Make this statement even stronger; it has to be emphasised ESM-Tools aid not in the coupling as such; just to avoid any misunderstanding
L131: Add the type of model and perhaps a reference.
L136: right -> access privelidges
L137: Very specific coupled setups are supported, right?
L138: What kind of conflict is meant?
L190ff: Thanks for including this paragraph, but perhaps revise a little to make it briefly more clear, how the user defined functions are embedded in with ESM-Tools.
L219ff: The listing of features is important and reads nicely, albeit I am wondering whether this is not a separate aspect, rather than a part of the Section 2.2.2 on the YAML configuration files; it reads rather like a section on "efficient maintenance"
L382: Perhaps mention some publications whose model runs were supported by ESM-Tools
L440ff: There is some redundancy and doubling of text it seems in the last two paragraphs of the discussion.

The line numbers refer to version 4 of the manuscript unless otherwise stated.

L59
both software package can be seen as natural
-> both software packages can be seen as natural


L66
"and have implemented first steps towards integrating full MESSy support into ESM-Tools, hoping to complete this process by release 5.0 (planned for October 2020)." This paper is supposed to already describe v5.0 now.

L108:
The ESM-Tools software is devided into three major parts
-> The ESM-Tools software is divided into three major parts
This was pointed out by the reviewer but not modified.


L172:
We separated our (Python)
I think the issue is to use present and not past tense


2.2.1 has lost a whole paragraph while the corrections by rev 1 are marked as done. Was intentional?

From reviewer #1
> P16L342 (in version 3 of the manuscript): ‘configuration’; is Done but ‘As an example, ...’ is not Done.
(in version 4: ...'the model. An an example, in Fig. 9,...')


Please correct the alignment of appendix subsections with the corresponding figures.
A1 Example ESM-Tools runscript
A2 Example YAML configuration file

It may be better to just use the subsection titles as captions and have an appendix without subsections.