We have had 2 good reviews of the paper and they differ in their view and what they focused on. I also thank the authors for their considered response and I also note some comments I made to the initial submission they also responded to.
The two core matters seem to be:
1) The tests are rather arbitrary, limited, and they are not comprehensive in nature - I also noted on the submission that they seem rather simplistic and not well justified in terms of any actual expected climatic variation - synthetic or not. I believe this needs to be resolved by improving the experimental base for the analyses, and I believe the authors intend to do this but they still need to be linked to some justification of 'climate change' else the paper should simply be about a sensitivity analyses and I am not sure that is enough for publication. So whilst the authors offer more % changes this is not enough to tie in better and more objectively with potential future scenarios and so the reasoning still has to be improved
2) That the model evaluation is improved - in the sense that some understanding of caution (or indeed not reported) for regions where the model does not do that well - which I add the authors have in some ways explored within their WRR paper. I believe I also noted this matter of model quality in my initial assessment.
The rest of the points on the paper are more minor and the authors have responded effectively to these
I look forward to seeing a more detailed analyses and revised paper that will be sent to the same reviewers again as they both expressed an interest in seeing the paper again, best wishes, Jim

Dear Authors,

The reviewers are now more favourable to the paper with all your edits and clarifications and this is appreciated. However there are some questions to respond to and please can you do some work on responding to those. If this is done to satisfaction (these are more minor edits now and relate to the tone and delivery of some messages in the paper) I will review the response and changes and decide if final publication is warranted. I do however want to note I do want you to pay special attention to any aspects that suggest, without any evidence being directly presented, that these stress tests relate to 'expected climate change". For example I see the follow quote is still in the revised manuscript (as an example). "The assumed changes of SSHIFT lie in the range of potential precipitation changes for winter and evapotranspiration changes in summer predicted by regional climate and water balance models for Germany until the end of the 21st century (Jacob et al., 2012, Herrmann et al., 2016, Paparrizos et al., 2018)" I keep on stressing this but you will have to provide direct evidence between your scenarios and these changes along with such statements else the reader cannot judge. So please either provide an appropriate analyses to show they are 'similar' to how you have formulated your stress tests or remove such comments...

Dear Authors, thank you very much for these final clarifications and changes to your manuscript. I believe the paper really well represents the experimental design you have achieved and the reviewers are happy with your final outcomes and discussion. I thank you on all your time in making these discussions useful to your paper and on top of that a number of my own comments. I believe we have a good contribution for HESS on a subject that is difficult to tackle and importantly challenging to provide the right context and interpretation. I am happy to have this paper sent for final publication to HESS, best wishes, Jim