Dear authors,

Thanks for your detailed online discussion replies. According to the referees’ comments the manuscript is generally suitable for HESS but still requires major revisions. Therefore, please prepare a revised manuscript and respond to all reviewer comments. The revised manuscript will then be re-reviewed by at least one reviewer.

I’d also like to add a few suggestions. As both reviewers criticized the negligence of parameter uncertainty, please take this comment seriously. Besides the suggested mentioning of previous studies in the revised ms, it would further be useful to add quantitative ranges on the amount of uncertainty that these studies found in order to fully respond to the referees concern over whether parameter uncertainty alone may be greater than other uncertainty ranges discussed/shown.

The NoCal run is apparently not entirely uncalibrated. Please be fully transparent on what exactly was tuned/calibrated/adjusted for this run.

The rebuttal to R2 3) is not entirely convincing because R2 attributes the lack of sensitivity to the fact that landcover parameters weren’t changed, whereas you reply that the lack of sensitivity is due to the abrupt land cover type change is minor. These are two different things unless it is made clear in the paper how both are related in the experiment. The concerns illustrate however that not everything is clear here and this then requires some clarification in the text.
Similarly some of the rebuttals to R1’s detailed questions don’t directly answer the posed question but turn to some related issue. Please review your rebuttal carefully in this respect.

I seem to remember that differences of WFD and WFDEI and their effects on global land surface modelling have been described in other publications. If this is indeed the case (and you likely know this better than I) please consider these in your interpretation and rebuttal and cite them.

New Fig. 3 The thickness of the lines in the graphs and the legend differs – needs to be the same. I suggest thicker in Figure as at present the line colors are very hard to distinguish.

A general comment: please refrain from ‘please note….’ sentences in the manuscript. If information on assumptions for parameterisation and other methods is needed then these should be stated comprehensively as such within the methods description and not be added ad hoc as a response to reviewer-requested clarifications.

Finally, I side with the referee who criticised that the abstract is not concise. This is a formal evaluation criterion for acceptance of a manuscript for HESS. Whereas HESS doesn’t have a word limit, the normal ballpark for an abstract is 250-300 words and yours is at least twice that and indeed not concise in its content. I am sure the senior co-authors can pitch in to make the abstract concise.
I am looking forward to reading the revised manuscript.

Best regards
Kerstin

Dear authors,

thanks a lot for submitting a carefully revised manuscript and detailed response letter.

While referee 1 and I now find the manuscript publishable generally, the readability could still be improved. Referee 1 made some specific suggestions in the re-review, which I ask you to consider seriously when preparing the final manuscript for upload into the system.

Best wishes,
Kerstin