Dear Authors

I thank you for responses to the review comments and the upload of an improved version already in response to those comments. I have had a detailed read over these, and it is apparent that many of the issues raised are resolved already through the discussion, and this is particularly the case for R1 and R2 comments. However, I do believe that the comments made by R3 are pertinent and reflect my own concerns about this paper in its current form. I have read your response to these concerns, but note they have not been thoroughly dealt with in the newly uploaded document, with only minor changes to the abstract, introduction and conclusion. I think the issues about the abstract/introduction/conclusion are not resolved and also clarity around the aims and use of the term data approach or data-driven approach or data-drive framework are essential to prevent readers like myself getting confused.

I therefore request that you make further major revisions to the paper to:

a) improve the definition of the scientific aims/objectives for doing this study - whilst I agree with some of your responses, the manuscript must more clearly define what the problem is that is motivating the research, more than general statements about numerical models being inadequate. What are the aims of the paper that make the approach and findings of interest to the HESS readership - what is the scientific question being studied? Currently the paper is focused on the specific case-study site making it hard to find the general relevance. You highlight the dataset as being unique and rare, but you still must have a clear question being addressed or methodological approach being assessed. Whislt you state the study has general relevance, you must more clearly prove that with specific detail. If it is the approach that is the significant contribution being made in this paper, then detailed discussion about the applicability and limitations of the approach/framework is necessary.

b) clarity in use and description of the approach being adopted. The term 'data approach' is vague and data-driven approach is ambiguous (relative to the common use of data-driven models in the HESS community for example). At some point the word framework is used. Can you be more specific in the abstract and introduction about the nature of the analysis? Maybe be consistent in calling it a framework and then outline the an overview of the framework in section 2.2? I'm not sure, but ideally it would be made more crystal clear upfront what the readers are going to be learning about and data approach is simply too broad.

c) I recommend detailed editing for grammar and style to improve the communication of your work. This applies throughout, but also includes rewriting of the abstract to follow a more typical structure (context, motivation/problem, aim, approach, results, outcome/significance), and the introduction (refer to R3 comments here). If you address the issues raised in points a and b then I hope this would make it easier. You must also proof read to make sure no basic mistakes exist in the revised version - I noticed many of them currently and it reduces the quality of the paper.

d) the discussion / conclusion are overlapping and need to be consolidated to directly address the aims and make clear the general relevance and limitations. I would suggest combining the Discussion/Conclusions into a single Discussion section, possibly with sub-headings to guide the reader about the flow of your ideas.

I look forward to seeing your revised paper as I feel these changes may take some more time, but will greatly strengthen the communication of your work. With your revision please make sure you refer to the comments by the reviewers, and summarise your modifications, or response to the comments. This will then go to a further round of review before a final decision.

Kind Regards
Matt

Dear authors

Thank you again for the resubmission and improvements to the paper. I have 3 new referee reports that have come in, and they have highlighted the technical suitability of the work, but still issues are present around the readability, style and presentation of the paper. In order to resolve better the requirements for publication I have myself undertaken essentially a 4th review with detailed recommendations - these are in the attached pdf with inserted comments (please let me know if you cant see the comments).

Reviewer 2 has indicated the paper should be more site specific, which is different from my previous advice. Currently I think the focus is OK, but I have recommended in my comments the inclusion of a diagram to better communicate the flow of the framework (data streams->analysis->outcomes/findings). This also will help solving issues I have raised in the attachment around confusion of data and analyses that are being referred to in the text.

Please also address other comments raised by the reviewer in your re-submission.

Dear Authors
Thanks for your updated revision. On reading this version I am glad to see the paper reads much more clearly and is better laid out. I think this is now a worthy paper suitable for publication and will be a useful contribution to the literature. I thank you again for you patience through the multiple revisions, and willingness to improve the paper.

Best regards
Matt Hipsey