This data paper serves as a nice bringing together of a variety of existing, but closely related datasets around the G2G national modelling. I find that having all of these details together in one place will be useful for a wide variety of UK-based research and modelling. I have tested all data links for the datasets presented and they currently work (2023-03-10). I have found the paper to be well written with only some minor comments below. 

I would note that I have selected good rather than excellent for originality and uniqueness because this paper is bringing together existing data rather than presenting new data. Though as above, I think this is still useful to bring these datasets under the same lens. 

Minor comments:

I haven't reviewed an ESS data paper before, but it seems odd to me to have what appears to be acknowledgements as the first paragraph of an introduction. Maybe for this paper format it is fine.

L98 - The authors may wish to note that the HadUK data used provides all the necessary variables to calculate PE if users require a higher resolution representation of PE. I understand that the intention of this data paper is to draw reader's attention to the soil moisture and flow datasets produced, but I think this addition would be useful.

L109 - I am missing some mention about why the convection permitting simulations weren't used. It seems this would overcome some of the steps needed in S2.3 and be more accurate in general. At least readers should be made aware of its existence. 

Figure 2 - the only lake cells I can identify are the 2 Northern Irish lakes. Are we supposed to be able to spot more? If so they will need more highlighting - or more explicit linking to the text about how the lakes are mainly significant in NI. Otherwise I'm not quite sure the point of Fig 2.

L238 - I think this paragraph can be written more clearly. Maybe it should start with "For the historical portion of the RCM PPE projections,..  "? But if so, it seems to overlap with the use of 'baseline periods' in the following paragraph (startin L251). I'm still a bit confused by it.

L267 - If my passing understanding of w@h and UKCP18 is correct (which it may not be), these two datasets can result in different (significantly different?) distributions of (e.g.,) precipitation, particularly at extremes. If this is true, it should be mentioned here. If there is no study that has made this comparison, then that is important information too.

Section 3.3 - It seems that the AMIN/AMAX soil moisture figure is missing here. It may not be such a conventionally studied metric but I feel is important to highlight the extremes. 

L416 - The authors may find it helpful to cite Schwalm et al. (2020) for this statement and thus support their choice of rcp8.5.  Also if I remember rightly there is only RCP8.5 for the UKCP18 regional projections anyway, which may also be worth mentioning and further justifies the use of this RCP.

Editorial:

L40 & L215 - hyperlinks don't seem to work

L324 - 'highly statistically significant increase' -> 'statistically significant increase'

References:

Schwalm, Christopher R., Spencer Glendon, and Philip B. Duffy. "RCP8. 5 tracks cumulative CO2 emissions." Proceedings of the National Academy of Sciences 117.33 (2020): 19656-19657.

The paper presents several datasets of UK river flows, soil moisture and derived statistics using both historical (CEH-GEAR precipitation, MORECS PET, HadUK-Grid min/max Temperature) and UKCP18 12km regional projection climate data. The hydrological model used for producing the datasets is the G2G model set up at a 1x1km spatial resolution and 15-min temporal resolution. The paper is generally well written. I only have a few technical comments as follows.

The river flow dataset provided is monthly mean, but the G2G model operates at a 15-min time step. Please explain why the daily river flow dataset is not provided. The daily flows can be more useful for the users.

The authors cited the performance of the G2G model in Lines 77-79. But it is not straightforward to find the performance of the catchments included in the datasets provided in this paper, because the GB and NI catchments were presented in separate papers and it is also not clear which catchments were included in this paper. I feel it is necessary to include the model performance of the selected catchments in this paper along with the datasets in a table format. This information should be added to Table 5.

Line 174-177:

The annual maximum/minimum flow values are nominally assigned to the start year of the 12-month period over which they are calculated, e.g. the annual maximum flow assigned to 1981 is for 1/10/1981–30/9/1982 (water years), while the annual minimum flow assigned to 1981 is for 1/12/1981–30/11/1982 (Dec–Nov years).

It is reasonable to use the two different water years here, but it is necessary to explain to the reader why the difference and if it affects the statistics of maxima and minima derived from the daily flows.

Please explain if the dataset of river flows is similar as the one presented in the other paper https://doi.org/10.5194/essd-2022-40 using the same G2G model. The latter contains daily flows from three models but only at the catchment outlets. Are the forcing inputs the same?

Figure 1 does not present much useful information as one cannot really tell the catchment area grids from the colours. I am unsure if this Figure should remain in the main paper.

Figure 2 is supposed to show majority lake cells >85% and >70%. I can clearly see the large lake (Lough Neagh) >85% but not sure where the area of lake cells>70% is. It might be the Lough Erne but not clear. It needs to be either noted in the text or highlighted in the map.