This is a review of the manuscript: ‘Implementation of Water Tracers in the Met Office Unified Model’. The manuscript describes the inclusion of water tracers in the UM model, as the title says. It provides the technical description on how the implementation was done, and an evaluation of the results of a 35-year long simulation. It is exciting to see that the UM Model now has implemented this possibility of water tracers, offering an online opportunity for moisture tracking alongside many existing offline moisture tracking models. I do wonder how the model performs for individual cases or tracking the sources of for example extreme precipitation events? This could be included as a possible outlook for future analysis?

The paper is well written and provides all detailed information on the implementation. I have some difficulties gasping the scaled-flux tracers description and the interpretation on the related figures 2, 5, 6 and 7. Further I have minor comments to clarify specific methodological steps, a suggestion for re-arranging the result section, and some suggestions to make the paper more accessible for (less technical) readers interested in water tracking. See all my comments below.

Comments:

Last sentence abstract: this sentence is quite vague to me (especially when you haven’t read the full paper), can you be more specific?

Line 43: There are also online models that embed numerical water tracers in regional models such as WRF-WVT (Insua-Costa & Miguez-Macho, 2018)

Line 45: One disadvantage of online water tracers if I understand correctly is that for tracking specific events you do not get the sources on a spatial grid but only per prescribed region (as the prescribed region tracer says), I think this should be mentioned here.

Line 67: at --> and?

Line 146: quantify the timestep

Line 150: the adjustment remains less than this level of 10^(-10) kg/kg at every..

Line 162: what do you mean with ‘normal water tracer’? is that referring to prescribed water tracers?

Line 190: indicate that the fields are functions

Line 200: So it is not possible to separate different land-use evapotranspiration sources? This might be a nice addition for future development

Prescribed regions; now chosen to distinguish ocean, land and ice, but this can be any region of interest? For example a country or river basin. I think it is important to make this clear in the paper that for further experiments this can be done.

Line 223: it is unclear what a ‘group’ exactly entails

Line 225-227. Twice ratio in one sentence but I think for the second ‘ratio’ the ratio of water tracer to water specific humidity is meant, but this is not entirely clear

Line 240: A three month spin-up seems very long for introducing water tracers, is this needed and can it be clarified/quantified?

Figure 2: Interpretation would be easier if latitudes (and longitudes) are added to the maps.

Line 249: I would like some guidance (in text) on how to interpret this plot on latitudes as it is not so easy to understand, and also comes back later in the manuscript. What can we learn from the plot (besides the comparison between the different tracers)

Line 269: numbers 6 and 7 --> should it be numbers 7 and 8?

Table 1: The results of water tracer number 21-23 are not shown in this manuscript right? Why include them in this table? And what is the (scientific) incentive to trace water given a wind speed limit?

Line 284: i_wt

Line 285: This argument counteracts with a spin-up time of 3 months?

Results: My suggestion would be to re-structure the results in order of sections; starting with section 3.2 as that is mostly model intercomparison and then figure 2 and figure 5 are more close together and can be connected more easily (as they show the same variable). Then afterwards the prescribed region tracers are evaluated in terms of general hydrological cycle characteristics. First discussing the percentages of precipitation source (now section 3.1) and then the hydrological cycle fluxes (now section 3.3). I think the section headers could also need some rethinking/rephrasing (the section header Prescribed region water tracers in the UM also works for section 3.3 for example).

Figure 3: mention what the black lines in the plots represent

Line 293: add evapotranspiration to clarify à of the precipitation sourced directly from land evapotranspiration (Fig. …)

Figure 5: same colorscheme as figure 2 as it shows the same variable? also extend the colorbar to minimum.

Figure 5 and 6 (and Figure 2); the caption says annual mean precipitation but from the text (line 314-315) I understand this should be ocean-sourced precipitation which I found confusing. Also if it is ocean-sourced precipitation why are there results over the land and ocean, then I would only expect information over the ocean? I have difficulties interpreting/understanding these figures/analyses given the current description (I know the scaled-flux is described in the methodology but still it is not fully clear to me).

Line 321: source temperatures --> source sea surface temperatures

Line 322: what is meant with ‘lower heights’?

Section 3.3: For me, the second Alinea on performance breaks the flow of this section and I would suggest to move the second Alinea to the end of the section and first compare the global hydrological cycle fluxes with the literature.

Figure 5 - caption: 30-year simulation --> in the abstract a 35-year simulation was mentioned?

Section 3.3: Hydrological flux results can also be compared to Demory et al. (2014)

Section 4: As this section also provides an outlook I suggest to change the name from Conclusions to Conclusions and Outlook

Line 394: between the two models

Line 403: 22% is still quite substantial in my opinion and is not really aligned with the word ‘only’ in front of the percentage

References

Demory, Marie-Estelle, et al. "The role of horizontal resolution in simulating drivers of the global hydrological cycle." Climate dynamics 42 (2014): 2201-2225.

Insua-Costa, Damián, and Gonzalo Miguez-Macho. "A new moisture tagging capability in the Weather Research and Forecasting model: Formulation, validation and application to the 2014 Great Lake-effect snowstorm." Earth System Dynamics 9.1 (2018): 167-185.