Review Comments:

This paper describes a simplified reservoir operating scheme that is distinctly oriented toward smaller non-irrigation reservoirs that only uses two parameters. This reservoir scheme uses general catchment and reservoir parameters that are more readily available than reservoir time series data. To test the sensitivity and performance of this reservoir scheme, the authors derived parameters for water supply reservoirs throughout Great Britain that encompass reservoirs of multiple sizes and uses. They also analyzed the feasibility of using transfer functions to create a nationally consistent parameterization. In their work, they evaluate the impact of two calibration methods one within each catchment and the other across the entirety of Great Britain.  Results show that the methodology works well in catchments with primarily water supply uses and is limited in catchments with multi-purpose reservoirs. This article does a really good job of creating simplified curves and evaluating metrics for evaluating reservoir dynamics with a unique focus on water supply reservoirs.

Major Edits:

• Equation 4 and 5 have Smin, which I assume refer to dead storage, however, this variable is not defined in the text. Please clarify this in the text so the reader is better able to follow what the equations are referring to.
• Section 3.5.1: As readers may be unfamiliar with the study domain, it could be useful to have a figure that shows the general makeup of the reservoirs used in this study colored by use, size and a second panel with the naturalized and non-naturalized catchments and general characteristics. This would allow a reader who is unfamiliar with the catchments in GB have a better idea of the characteristics of the catchments mentioned in the paper.
• Line 308: I am personally not familiar with the catchments in GB. To increase clarity, I would include a catchment map either with numbers or highlight these catchments in Figure 2 so that the reader knows which area you are referring to. Alternatively, you could leave out the reference to these specific catchments and describe them more generally. The same also goes for Line 388, 390, and 460. I would make the distinction as to why you picked these catchments and give an overview of them prior to mentioning them so the reader is not confused.
• Line 313: This is an interesting note about reservoir construction in GB and it’s impacts on your work. What would be the larger impacts on other regions that are more groundwater regions dominated such as the southwestern US or for dams built on limestone or more porous rock such as the Mosul Dam in Iraq?
• Line 337: The rational for using a linear relationship is not clear from the text. Why was a linear relationship better than a non-linear one?
• Line 350: Including Table 1 with the upper and lower bounds of the transfer function parameters is nice, however, the Table is not referenced at all in the text. Instead of a table, it might be nice to include a figure of the parameter ranges, average parameter value, or another metric to show the regional differences in transfer function parameters across different catchments in GB.
• Figure 4 and Figure 5: Why did you pick the four catchments denoted by stars in these figures? What characteristics cause those to be chosen for the case studies and do these directly relate to the ones mentioned in Line 308, 388 and 390? If so, I would make the connection between these clearer.
• Line 549: Can you expand on what you would suggest other do in cases where your methodology is limited (i.e. in the case of multipurpose reservoirs)? Should a combination of methodologies be used or a more generalized approach such as Hanasaki et al., 2006? Additionally, how would the authors envision upscaling this approach to other hydrologic models?
• Section 4: The authors do a really nice job of explaining their methodology and results. It would be interesting to see a comparison between this simpler model and one or the more generalized approaches or perhaps the approach that is already in DECIPHeR as another point of evaluation.

Minor Edits:

• Line 53 and 55: I believe the abbreviation for GRanD should have a capital R as well.
• Line 60: ResOpsUS contains over 600 dams. I would edit this line accordingly.
• Line 125 – 128: I would suggest adding parenthesis to equation 4 and 5 to denote the right side of the comparison.
• Figure 3: I would switch units in this figure from Ml/day to cubic meters per day since most audiences are more familiar with cubic meters.
• Figure 6: The color difference between reservoir and observed is hard to distinguish. I would suggest altering the colors to be a bit more clear. I would also change the units of CF and ABS to be cubic meters /sec or per day so the units across all graphs are consistent. I do like that you kept the colors of the stars the same throughout Figure 4, 5 and 6.
• I like the colors on Figure 7 and the difference between the three curves is much easier to see. I would also change the units on CF and ABS to cubic meters per second or day for consistency.
• Line 570: Could you give an example of a signature or test that you would suggest for other studies?
• Line 589 – 590: Perhaps I missed a section, but I am not sure where the first half of this sentence comes from since the large scale hydrologic models that I am aware of all contain reservoirs in Great Britain. If the authors are referring to water supply reservoirs specifically, I would suggest rephrasing.

General comments

This paper reports on the development of a generic algorithm for water supply dams and its application to a hydrological model for the United Kingdom. The development of hydrological models that include reservoir operations is essential for estimating flow rates and water resources. The paper argues that a relatively simple model can represent the operation of water supply dams in the UK. The results reported are generally considered reasonable, but some of the descriptions were difficult to understand and should be refined.

Specific comments

Line 185, ”three equal classes of slope and accumulated area”: This part is hard to read. What do you mean by “equal classes”? What is “accumulated area”?

Line 218, “to determine unbiased values for the natural model parameters in reservoir catchment”: What do you mean by “unbiased”?

Line 234, “the OS river layer”: What is this?

Line 252, “5000 times in the reservoir scenario sampling reservoir parameters”: I couldn’t understand these runs and how they were implemented. What do you mean by “sampling”?

Line 264, “In this study abstraction and compensation flow remain constant throughout the simulation.” This is an interesting point. It implies that the summation of the abstraction and the compensation flow must be equal to the mean annual inflow (otherwise, the long-term water balance does not close). The key difference between this study and Hanasaki et al. (2006) is whether to separate the outflow of reservoirs into abstraction and compensation flow. This point would be mentioned somewhere.

Line 304, “The results from the near-natural simulations for the best nationally-consistent set of transfer function parameters”: It is a bit awkward to read this. Why don’t you put a simulation name? I think the combination must be [NATural or REServoir], [INDividual or national-CONsisnt]; hence, this run might be called “NAT-CON.”

Line 315 “(or simulations, decided based on those with the highest median non-parametric KGE across all 137 catchments)”: Hard to read. Better to clearly explain/define this in the Methods section.

Line 333, “to define the transfer functions used in this study”: It is a bit confusing. I believe at least two transfer functions are used in this study: one for hydrological parameters and the other for reservoirs. For better readability, these should be clearly distinguished.

Figure 3: Please confirm whether the label “Top national simulation” is correct. I expected “Best national simulation”. Anyway, the mixture of top and best is confusing for me.

Line 338-339 (Equations 6-7): Again, this is a very interesting point. The compensation flow is a function of the catchment area. Because the inflow to a reservoir is basically proportional to the catchment area, it can be said that this equation is similar to Hanasaki et al. (2006; CF ~ mean annual inflow). The abstraction is a function of reservoir capacity. Because the reservoir capacity is only weakly correlated with the inflow usually, it can be said that this formulation is different from Hanasaki et al. (2006). Then, let me come back to my previous point. The equations 6-7 do not seem to guarantee the water balance shown in Equation 1. From this perspective, it is doubtful that the overall modeling framework is hydrologically reasonable. This point should be additionally discussed in the manuscript.

Line 341 “is in line with the observed data”: What does it mean by “in line with”? The gray dotted line is basically above the blue dots.

Line 343 “at the upper end of the sampling limits”: What are the “sampling limits?”

Line 343 “the range of variability of the transfer functions associated with the top 5% of national-consistent simulations”: Top 5% in terms of what? What do you mean by “the range of variability”?

Line 376 (Figure 4 caption) “top reservoir simulation”: Again, what is the difference between “top” and “best”?

Line 502- “the simplicity of our operating rules”: Ideally, the author should also demonstrate that their model outperforms the even simpler model (Hanasaki et al. (2006), which requires no calibration parameter).