Dear Author,

Thank you for submitting a revised version of your manuscript submitted to Biogeosciences, which can be accepted for publication after minor revision. When submitting the revised version, please let me know which changes were not implemented, if any, and why. This will speed up final acceptance.

I look forward to seeing this paper published and thank you for considering Biogeosciences to publish these very interesting results.

Best regards,
Jean-Pierre Gattuso, BG Editor


-----------------------------------------------------------------------------
- CT and AT (C and A in italics, T as an upright subscript) are the preferred symbols of dissolved inorganic carbon and total alkalinity
- limit the use of acronyms. For example, CCA and OA could often be spelled out, improving the reading experience.
- I already raised the issue of how the models are referred to but my comment was misunderstood. You use superscripts attached to the name of the first author or of a model, making a very awkward definition and cumbersome reading experience. I do not think that I cam across something like this before. Most often, an acronym is made using the author names. For example model of Berner et al. (1983) is referred to as BLAG (Berner, Lasaga and Garrels).
- 143: G in italics
- 84: what do you mean by "sensu Erez"? I am not sure what is meant because this is a well established fact.
- 86: I think that it is a pity to forget about early papers. The alkalinity anomaly technique dates back from the 1970s and was first used in coral reef settings by Steve Smith. In this context, it is a little misleading to cite a paper published in 2013.
- 88: z is water depth
- 88 forward: it is worth pointing out that the AT technique is most often used to measure G of a portion of reef (reef flat) rather than the whole reef ecosystem.
- 97: define E
- Superscript: inconsistency for irradiance which is referred to as E in the text (which is correct) and to Irr in the model definition.
- 98: scaling up future changes in G also require changes in the community composition
- 118: Biosphere 2
- 122: incorrect underlying assumptions
- 128: terms?
- 153: hyperbolic tangent
- 155: Ez is not the light penetration; it is the irradiance at the depth considered.
- 157: the preferred unit for irradiance is umol photons m-2 s-1
- 161: you need to explain what K490 is
- 159: Ec is not defined. Should it read Ek?
- What is the unit of Greef?
- 203: delete "relative"
- 213: what is the unit of Gcoral?
- 215 using the alkalinity anomaly technique...
- 223 and 224: ' signs are different in the equation and text
- Topt does not seem to be defined
- 232 and elsewhere: always add a non-breakable space between a number and its unit
- 236: salinity is defined as the ratio of two conductivities and therefore unit-less
- 274: was the value of live coral cover really reported in all the studies that you considered?
- 292 and elsewhere : there should be spaces before and after ± (+-) signs
- 297: low CALCIUM carbonate
- 313: what do you mean by "equitable latitude"?
- 362: which is almost half Smith's estimate
- 398: degree of LIGHT attenuation
- 470: genus rather than genera
- 480: Note that in my review published in 1999, I found that dissolution of individual corals has very rarely be reported and that the median ratio of light to dark calcification is 3.0.
- 517: irradiance rather than light intensity
- 523: the increase in calcification as a function of increased temperature obviously stops at a certain thereshold.
- 557: mortality
- The quality of Fig. 7 does not seem to be very good. You should perhaps decrease the thickness of the lines.

Dear Author,

Thank you for submitting a revised version of your manuscript submitted to Biogeosciences, which I am glad to accept for publication.

I look forward to seeing this paper published and thank you for considering Biogeosciences to publish these very interesting results.

Best regards,
Jean-Pierre Gattuso
BG Editor