Dear authors,

three reviewers have given feedback on your manuscript. The reviewers give generally very positive feedbacks and state they were intrigued by the analysis and results. All find that it is a timely and valuable contribution to the field of global hydrology. The paper is well structured and written. The reviewers have also given constructive feedback and criticism and you have addressed several of those comments in your response.

I agree with the assessment of the reviewers on the merit and novelty of the presented analysis. I would however like to emphasize one point: All of the reviewers comment, in one way or the other, on the fact that the results hinge upon the correctness of the CDR dataset. In your response you emphasize how carefully the CDR dataset was developed and validated, and also your own efforts to validate e.g. the standard deviation of E. I appreciate this. However, some of the standard variations in the dataset are not yet validated. I agree with reviewer #2 (René Orth) that a cross-validation would be desirable to learn whether the observed variance patterns are a property of the CDR dataset or hold with other datasets. At the very least, and since the main message of the manuscript is a call for investigation into the causes of the observed hydroclimatic variability, the discussion should more than now acknowledge to that fact that any efforts towards validation of those patterns are equally warranted.

Please submit the revised manuscript, with changes highlighted, together with a point by point response to all of the reviewers comments.

I thank both the authors and reviewers for the constructive discussion and look forward to the revised manuscript,
Anke Hildebrandt

Dear authors,

thank you very much for the revised version of your manuscript. Since all other reviewers suggested minor revisions, I only requested one of the reviewers (René Orth) to comment on the new version.

He appreciates the additional analyses performed, but still feels that some of the variables have not been validated, such as run-off. The reviewer suggests additional analyses, regarding (1) use updated version of the Jung et al. dataset (Jung et al., 2019) and (2) use the E-RUN dataset (Gudmundsson and Seneviratne, 2016) to validate the runoff.

I find that the manuscript has much improved and you have made good effort addressing the reviewers comments. I think the manuscript is almost ready for publication, given some amendments. In view of the fact that the Jung et al. (2019) paper was published only after the submission of the manuscript (although the data were available earlier), I will not insist on this additional analysis. However, please consider the Gudmundson and Seneviratne (2016) dataset. Please also attend to the detailed comments of the reviewer.

I am looking forward to your resubmitted manuscript,
Anke Hildebrandt

Dear authors,

thank you very much for the revised version of your manuscript, including the comparisons with additional data. I find the manuscript is now ready for publication. Thank you for this thought-provoking contribution to HESS!

Best regards,
Anke Hildebrandt