General comments

The overall quality of the preprint is good, and the described developments of the S2P3-R model are suitable for publication in this journal.  The manuscript describes novel model updates and evaluations of the model both globally and regionally.  The information and data provided will allow others to assess if the model may be appropriate for their use.

Specific comments

No comparison is made of regional biogeochemical performance compared to a long reanalysis such as the Copernicus Marine Service NORTHWESTSHELF_REANALYSIS_BIO_004_011 product.  This seems like a missed opportunity given the spatial coverage compared to satellite data.

Suggested minor revisions

Comparisons to satellite data state data was limited to “case 2 water, i.e. water ≥ 70m water depth (Jackson et al., 2019)”.  Whether this is meant to be “case 1 water” or “≤ 70” isn’t clear as the reference “Jackson et al 2019” does not seem to be available.  I would suggest a clarification of satellite data selection criteria.

The data variability in Figure 9A and 9B (and to a lesser extent Figure 18) is difficult to distinguish with a grey background. I would suggest the use of a white background for plots with viridis colourmap, such as in Figure 7. These global plots would also benefit from being larger, single viridis and blue-white-red colour bars could be positioned either side.

Figure 13 would benefit from enlargement and using a log scale may be more appropriate.

Technical corrections

Page 12 Line 21: suggest replacing “other” with “apart from”

Page 29 line 41: This reference doesn’t appear to be available from the url provided.  

General Comments:

The paper presents an upgrade to a previously published modelling system. Addressing forcing issues that allow the system to be run over larger areas and longer times. I believe this to be a sufficiently large advance in modelling science to merit publication. The methods are clear and well presented. The protocols appear to be well documented with the supplied code (though I have not tested them).

The document presents an honest accounting for the strengths and weaknesses of the modelling system. In places this is a little too sweeping, or lacking the detail that would permit the reader to make scientific inferences from the results. E.g. to what extent does the exhibited skill over the Patagonian or North West European Shelf imply that later fluxes are not important. But that is not the aim of the paper.

The results support the concluding remarks, except that I would more strongly state the possible value of this tool in education (perhaps to undergraduates?). I also think that the value to policy groups of this “cheap” model is perhaps slightly dangerous if the output are not in some way corroborated with with existing data from higher-expense simulations. After all this paper, at length, highlights the gains in efficiency do come with a loss in skill.

Specific Comments:

p9. Fig 3 caption: line 6: Without parenthetic commas, the  “therefore” comes in the wrong place. E.g. This might be clearer:

“Where this is positive there is a net heat flux into the ocean. So, assuming the system is approximately at steady state, advection of heat is therefore out of this area.”

p9 line 15: should read “… more prevalent at low M2 tidal amplitudes…”

M2 is dominant in the North West European Shelf, in most places. But K1 can be relatively large in other regions, like the South China Sea

p10. Line 2. Unpack this line. Is it the case? Does Figure 5 exhibit smaller model biases in the summer? Confirm what you think my eyes are telling me.