Thank you for your responses to the referees and your revised manuscript. I have now reviewed these and am pleased to report that I am satisfied that you have addressed the referees’ concerns.

That said, one of the issues raised by Referee 2, namely their concern about all of the NPP going to the modelled commercial species, could be dealt with more convincingly. While the fraction reaching the modelled groups is perhaps not well-known, it is at least known that it’s not the 100% assumed here. As the referee remarks, the plethora of non-commercial species, as well as export to the deep ocean, represent a sink for NPP that BOATS overlooks. My interpretation of your response to this point is that since your main results scale linearly with the fraction assumed, you exclude this parameter from your analysis in order to simplify it. This is sort of fine as far as it goes (i.e. a sensible short-cut for model world), but as the referee notes, does it not perhaps overlook commercial fish biomass as a data constraint? That is, if you choose different values, you will have more / less biomass and will conform to reality more / less. Perhaps, however, there is insufficient data to do so (you do also note “the scarcity of appropriate data constraints”).

In passing, the concluding sentence in this section (“By assuming that ...”) is something of a non-sequitur to the preceding sentence.

Anyway, assuming that my interpretation above is accurate, I think it would be helpful for the text to more clearly note that this assumption is made purely to facilitate analysis rather than being what you actually think the fraction is. And on the data constraint point, perhaps noting the uncertainty range of biomasses might convince readers not to get overly concerned about this assumption.

If this final point can be cleared up, I see no problem in accepting this manuscript. Thanks again for your efforts, and for dealing so completely and clearly with the referee concerns.