I thank the authors for responding in detail to the two reviews and my own comments.

Despite of these responses, three significant weaknesses remain:

1) The authors have not adequately responded to the transboundary nature of the subject matter, and the concerns expressed by reviewer #1 (point 3). The authors’ rebuttal (“Since GWM try to capture only the uncommitted monsoon flows, it will not affect the downstream riparian regions including Bangladesh. Rather it will have positive impacts for Bangladesh in terms of mitigation of floods in the monsoon, and potentially more flows in the river in non-monsoon months in the long-run.”) is not at all convincing.

a. First, water demand, and in particular “unmet water demand” in Bangladesh is simply ignored; except for the fact that environmental flow requirements (EFR) are discussed (see e.g. how “committed outflow” is defined); indeed EFR could partially meet such downstream demands, but this fact is not even mentioned. This disregard for downstream water demand may be interpreted by some as denying legitimacy to downstream water demands and even water rights.

b. Second, heavily investing in SSS in upstream catchments and subsequent increase of the consumptive water use there will surely have significant downstream impacts. In particular the so-called ”preparatory pumping” from groundwater before the monsoon may have far reaching impacts for dry season flows downstream. This issue may not be glossed over, and should be explicitly addressed.

2) My own earlier comment on existing over-exploitation of groundwater in large parts of India including in parts of the Ganges (which is well documented in the scientific literature) and how this relates to the current manuscript is in my view not adequately addressed.

What I want to see in e.g. Figure 8, is in which parts of the basin there is already over-exploitation of groundwater, and where there is not. Further, where there is currently over-exploitation of groundwater, I would like the authors to assess whether there is still “potential unmet demand” (i.e. the volume/flow is still positive) or whether current uses are beyond that threshold (and the amount of “potential unmet demand” should be a negative number). What criteria are to be used to make this distitnction? Finally, where there is over-exploitation of groundwater, the authors should reflect to which degree there is already a kind of “Ganges Water Machine“ going on in those areas, if not, why not, if so, how, with what impacts and what can be learned.

3) I feel that the paper does not provide evidence for the following conclusions as formulated in the very last paragraph:

“However, the benefits of developing only a small portion of such potential storage can be enormous, because SSS can enhance sustainable water supplies, and provide benefits for irrigation and other water-use sectors. It can buffer rainfall variability and reduce extreme flooding, especially in downstream regions. SSS can increase river flow during the low-flow months either through baseflow or reallocation of canal irrigation. Importantly, it can mitigate the negative effects of floods and water scarcity in the same year, which often affects the most vulnerable people of society – women and children, the poor and other disadvantaged social groups.” (p. 8745, lines 3-10)

These are assertions that may be plausible, but have not been proven in the paper, and actually echo the same assertions made in the introduction (e.g. p. 8731 lines 1-8). So the progress achieved by this paper is disappointing.


Finally, I find the title of the paper to general and little informative. The fact that the paper (only) deals with the demand-side should be reflected.

I encourage the authors to carefully consider these comments and re-submit a revised version that adequately addresses the above concerns as well as those articulated by the two reviewers.

Dear authors,

Thank you for submitting a revised version, and for cearly indicating how the reviewers and editor comments were addressed and which changes made.

I note this is my third comment to you the authors.

I am of the opinion that the ms has been significantly improved, mainly by (1) toning down the claims, (2) including water use in the Bangladesh part of the basin, and (3) making some spatially explicit nuances as to where the unmet demand might possibly be me and where not.

There are still some weaknesses that need to be addressed, but these are, I think, feasible.

1. The first comment I made in my initial comment, namely the need for a precise definition of the “Ganges water Machine”, has not been addressed (at least I couldn't see it). Can yuo please somehwehre clearly mark the definition used.

2. Another comment in that initial comment, namely about the sloppy use of units, is still an issue. I have indicated some instances where this must be improved in the attached marked up pdf.

3. The spatially nuancing of where the unmet water demand could be met should have consequences for the total amounts mentioned in the abstract and the conclusions (59 and 124 Bm3). Perhaps Table 4 should be amended to indicate more clearly where to what extent the potential increase in groundwater CWU can be met or not. This is indicated in the text, but not in the table. It would be good that a total feasible value could appear, both in the table and in the text.

4. The new figures 8a and b are not clear (have a bad resolution in my copy)

5. There are still several minor editorial issues, as indicated in the marked up pdf (attached)

Success!

Pieter van der Zaag

Dear authors,

Thanks for adequately attending to the final touches of the paper. I recommend the paper to be accepted for publication.

Pieter van der Zaag