Dear authors,
This has not been an easy ride but I am happy to let you know that the two reviewers agree that you have met with all their questions and suggestions. The reviewers stressed that some parts of your texts seem very long and with complicated sentences but that the observations are described objectively, the reasonings are all scientifically sound, and the conclusions drawn are based on the presented data. Therefore, your manuscript can now be accepted for publication. Congratulations!
Best wishes,
Jan-Berend Stuut

Dear author,
I am glad to accept publication of a revised manuscript pending the incorporation of suggestions by the two reviewers and a few comments below (these concerns the first submitted version):

section 4.1 p. 26: The comparison with nutrient concentrations from other studies could be shortened (seems like the authors need to "defend" their results, although at these low concentrations interannual variability would be expected).
lines 544-546: why use a POC/Chla ratio from the literature, when (I assume) you have your own POC measurements for the tanks.
Lines 569-600: this should be shortened (focused on this and not previous or future work) and possibly shifted to the results.
Line 601-612: this can also be shortened and possibly shifted to materials and methods.
Lines 613-631: this can also be shortened and possibly shifted to section 4.4
Lines 643-654: could also be shortened and focused on results presented here, rather than the argument based on the comparison with a previous study on dry deposition.
Lines 689-698: this can be removed. The data suggests strong grazing control by nanoheterotrophs in particular at TYR tank G1, so I would also be careful with the statement Line 688 "complete absence of response from autotrophs" (see also Fig. 7) and also lines 834-836.
Line 719: why do you assume competition for macronutrients here? NOx is clearly in excess of demand, DIP decreases faster than the increase in heterotrophs, and grazers are dominated by nano-eukaryotes that feed on particles not dissolved matter.
Lines 731-736: I understand that the argument here is that the reproductive cycle of metazooplankton is longer than the incubation times (agreed), however, the increase in zooplankton is mostly due to the lumped Harosa/SAR counts (Fig. 9). These include protists that can grow fast. Is there additional info on the assemblage composition within this group?
Figure 8.: Please specify that the counts in the two upper panels correspond to autotrophic eukaryotes (isn't it?).
Lines 834-836: This statement contradicts data shown in Fig. 10 (see also previous comment on lines 689-698). Figure 10 gives a good overview of differences between treatments. I believe the manuscript could be improved by discussing Fig. 10 with the other results in section 4.4. This would also help shorten the conclusions to the main take-home messages.
Sincerely,
Christine Klaas

Many thanks for your excellent contribution to the special volume and congratulations on a very interesting paper!
It has been a long trip but I hope you agree that the result is very good.
Cheers,
Jan-Berend

Dear Authors,
I am happy to accept publication of your manuscript pending minor revisions: the comments from the reviewers were well attended to. I believe, however, the text can be improved and somewhat shortened avoiding unnecessary details and repetitions. Before submission I would also recommend a thorough check of figures, tables and their legends. Comments are given on the annotated manuscript.
Sincerely,
Christine