Please make the code available, then describe how it may be obtained. Please also provide the DOI for the exact version of the code described in the paper.

Other materials may be uploaded to a public repository with a DOI, or added to the supplement. Weblinks are very ephemeral. Try to include as few weblinks as possible in the final version of the paper.

This paper and its companion paper are fairly epic. They could really do with clickable tables of contents, but unfortunately that is not supported within the journal formal. I am very grateful to the 4 reviewers for their carefully considered reviews.

For such long papers in GMD it is important that as many of the details are correct as possible. Readers may well dip into the papers, and make use of small details. If there are errors then this considerably reduces the utility of the paper. At the same time, it is unlikely that any of the peer review team are in a position to find these kind of errors. So, you can either check the paper very carefully yourselves at this stage, or you could also ask people who know the model well but are not authors on the paper to check the content for errors. It is up to you, but the latter method has been employed with considerable success (i.e. some important errors were found!) for other long papers in GMD.

As discussed over email, please update the paper with corrected information, including DOIs for the code and data that are being made available.