Dear authors,
I have gone over your responses to the reviewers' comments and now invite you to prepare and submit a revised version of your manuscript. Please provide a version were modifications are visible i.e. track change mode, and another one in which they are accepted. They will be sent for additional review to the original referees.
Thanks you for submitting your work to Biogeosciences,
Best regards,
Sophie Bonnet

Dear authors,
Your manuscript has been sent to the original reviewers, who find that the manuscript has greatly improved. However one of them still recommends minor revisions (see below) before your manuscript will be accepted.
Best regards,
Sophie Bonnet

This revision is much improved – the more detailed descriptions of calculations and associated caveats are appreciated. I have a few comments on minor points that could use further clarification.

You might add a supplemental table with the actual statistical results, PDF document would be fine, just to make that info available for those who are interested. Likewise, will the actual data be archived somewhere? If so, you should point the readers to that repository; if not, you might consider putting the results in a supplemental file (i.e., concentrations at each depth at each station, in much more detail than Table 1). A major advantage of this work is that it contributes to our knowledge of an under-sampled region of the ocean, so those results will certainly be of use to the broader community.


Line 26: Please say 18degS instead of -18degN.

Line 29: Are you saying there’s no difference in DOC concentrations at specific depths? Because at first glance this statement reads as inconsistent with the differences in integrated stocks reported just below. You do need to be careful throughout the manuscript to be clear when you’re discussing concentrations (no differences between regions) vs. integrated stocks (significant differences between regions). E.g., Table 1 vs. Figure 6 – the distinction is clear with attention to detail, but I think it’d be worthwhile to make sure your readers have minimal opportunities to misinterpret. The body of the text is generally fine, it’s more in the figures, abstract, conclusions, etc.

Line 125: This link doesn’t work for me (it may be temporary, not sure). Might be less problematic to replace with a citation to something from Hansell’s group talking about the ref waters.

Line 300: This is confusing and needs to be rephrased – you say you’re calculating the residence time of DOCsl as DOCex, but then you talk about how the calculation overestimates the residence time of ultra-labile DOC. Are you trying to say that the residence time is over-estimated *because* it’s calculated using uptake of ultra-labile DOC?

Line 353: biogeochemical rather than biochemical?

Line 365: This raises the question in my mind of whether the MA area experiences higher turnover or seasonal mixing than the gyre, to the extent that it would compensate for the shorter residence time of DCNSs in that region, such that the export might not be that different between the two regions, just controlled by different mechanisms? I am definitely not insisting that you go down that path, but it could be interesting.

Line 372-373: Can you cite something for the statement that exopolysaccharides would be hydrolyzed (suggesting the production of nitrogen-requiring exoenzymes) but then not taken up? Or if speculative, please be more explicit that that is the case. Or are you saying that glucose is a by-product of other exohydrolytic activity that isn’t used and so accumulates? If the latter, this sentence needs rewording – took me a while to get to that possible meaning.

Line 385: missing an adjective (higher/greater)

Lines 395-396: I would remove this sentence – it’s not wrong, but it’s not a meaningful conclusion from this particular study.

Figure 4: The relatively high deep-water DCNS concentrations at ~175W are intriguing – is there a water mass difference that could be contributing? Quite a lot higher than the values you cite in Line 218.

Figure 6: Perhaps take a look at an axis break between the DOCex and DCNS-C values, to make DCNS-C easier to see, if differences or lack thereof between regions is the point? Unless the point of the figure is to emphasize that DCNS-C is a very small percentage of the overall DOC stock, in which case it works well for that purpose.

Throughout: Mann-Whitney is written as Man-Whitney in several places.