The authors projected the climate change-related impacts in the marine ecosystems of the Mediterranean Sea in the middle and at the end of the 21st century using an offline coupling model combining the physical model MFS16 and the transport-reaction model OGSTM-BFM, under emission scenarios RCP4.5 and RCP8.5, focusing on the middle and the end of 21^st century. Projected changes are presented for temperature, salinity, dissolved nutrients and oxygen, net primary production, respiration, organic matter, plankton and bacterial biomass, particulate organic matter, and biogeochemical parameters (DIC, pH).

The paper provides interesting projections in a changing Mediterranean Sea that is already under multiple pressures. Please check below my major and minor comments:

Major comments:

P3, L75: No, not “all” the modelling studies focused on high emissions scenarios. For example, there is Benedetti et al. (2018) who used A2, A1B and B1, and Goyet et al. (2016) who used B1 and A1F1.

• Benedetti, F., Guilhaumon, F., Adloff, F. and Ayata, S.-D. (2018), Investigating uncertainties in zooplankton composition shifts under climate change scenarios in the Mediterranean Sea. Ecography, 41: 345-360. https://doi.org/10.1111/ecog.02434
• Goyet, C., Hassoun, A., Gemayel, E., Touratier, F., Abboud-Abi Saab, M. and Guglielmi, V., 2016. Thermodynamic forecasts of the mediterranean sea acidification. Mediterranean Marine Science, 17(2), pp.508-518. Thermodynamic Forecasts of the Mediterranean Sea Acidification | GOYET | Mediterranean Marine Science (ekt.gr).

Goyet et al. (2016) is the only modelling study that it is projecting carbonate system parameters in the Mediterranean Sea so far, and the one used in MedECC (2020 ; cited by the authors to tackle OA projections in the Mediterranean). Yet, it is not mentioned at all in this work. Please check this study out and try to compare your results with theirs.

P3, L78-79: The authors mentioned that Moullec et al. (2019), under RCP8.5 emission scenario, found an increase in both phytoplankton biomass and net primary production by the end of the 21^st century. However, this pattern is not homogenous in the Mediterranean since Moullec et al. (2019) have also highlighted a difference between the Eastern and Western basins with an increase in the first and a decrease in the second. Please edit accordingly.

P4, L111-113: In addition to the BOUM mesoscale experiments working on relating eddies with biogeochemical changes (BG - Influence of anticyclonic eddies on the Biogeochemistry from the Oligotrophic to the Ultraoligotrophic Mediterranean (BOUM cruise) (copernicus.org)), there are actually many modelling studies, for example:

• Ramirez-Romero E, Jordà G, Amores A, Kay S, Segura-Noguera M, Macias DM, Maynou F, Sabatés A and Catalán IA (2020) Assessment of the Skill of Coupled Physical–Biogeochemical Models in the NW Mediterranean. Front. Mar. Sci. 7:497. doi: 10.3389/fmars.2020.00497
• Guyennon, A., Baklouti, M., Diaz, F., Palmieri, J., Beuvier, J., Lebaupin-Brossier, C., Arsouze, T., Béranger, K., Dutay, J.-C., and Moutin, T.: New insights into the organic carbon export in the Mediterranean Sea from 3-D modeling, Biogeosciences, 12, 7025–7046, https://doi.org/10.5194/bg-12-7025-2015, 2015.
• Herrmann, M., Somot, S., Sevault, F., Estournel, C., and Déqué, M. (2008), Modeling the deep convection in the northwestern Mediterranean Sea using an eddy-permitting and an eddy-resolving model: Case study of winter 1986–1987, J. Geophys. Res., 113, C04011, doi:10.1029/2006JC003991.

Therefore, I would suggest to re-write this paragraph.

P8, L263-266: To characterize the spatial distribution and the variability of anomalies, the authors considered their horizontally averages in each sub-basin in the Western Mediterranean (WMED=(ALB+SWM+NWM+TYR)/4) and in only two sub-basins of the Eastern Mediterranean (EMED=(ION+LEV)/2). Why did you exclude the Adriatic and the Aegean Sub-basins here?

P9-10, L306-315 & Fig. 2: The authors mentioned that mean simulated values in the first 0-200 m are quite realistic in all the variables, and that biases started to show at 600 m depth. However, the vertical profiles show such discrepancies between CTRL average profiles and observational data (EMODnet) even in shallower depths, i.e. less than 50 m for phosphate in the WMed., surface waters for nitrate in the WMed., greater than 200 m for oxygen, and so much general biases in pH. Could you please elaborate more on this?

P11, L326: Could you explain in the text the depth classification adopted in this study: 0-100 m and 200-600 m?

P12, L342-358: Is it possible to check if those differences are significant or not?

P20, L476: Please explain briefly the role of “damping effect” in controlling oxygen values at the Gibraltar Strait?

P24, L565-566: I guess you are talking here about the “projected” change not the “observed” change. In any case, I would suggest to better re-write this sentence.

P29, Fig.16: Captions on the plots should be corrected to distinguish between its different components (a-f), as well as between the locations (Med., WMed., EMed.).

In section 2, authors refer to alkalinity (ALK) in the text (i.e. L188). Do you mean by this term, the number of moles of hydrogen ions equivalent to the excess of proton acceptors (bases formed by weak acids) over proton donors (acids) in a kilogram of sample? Mostly yes, and this term should be labeled total alkalinity (DOE, 1994): TA = [HCO3-] + 2[CO32-] + [B(OH)4-] + [OH-] – [H+]. Moreover, except for figure 2 for the period 2005-2020, there is no results about ALK in the following sections. Why?

The term “tracers” is usually used for conservative elements that can be traced in function of time. It is not the appropriate term for the carbonate system parameters, such as TA, DIC or pH. Please refer to them as biogeochemical parameters/features/properties but not tracers.

Minor comments:

Please write the E and W in Eastern and Western in capital letters, and unify this in the text.

While I would suggest to add “sub-basin” to any sub-entity in the Mediterranean (i.e. Alboran Sub-basin, Levantine Sub-basin, etc.), it is OK to use “sea” instead like many other publications (i.e. Adriatic Sea, Aegean Sea). However, I would recommend the authors to unify the terms adopted throughout the manuscript since they use “Adriatic Sea and Levantine basin”, why? Also, sometimes you refer to the Gulf of Lion as Gulf of Lions (i.e. L445). Please rectify and unify this in the ms.

Please make italic the “a” in Chl.a throughout the text.

Please write “time-series” instead of “timeseries” throughout the entire manuscript (as you have already done it in L502).

Abstract:

P1, L16-18: Please write it as follows “The analysis shows significant changes in the dissolved nutrient content of the euphotic and intermediate layers of the basin, of the net primary production, phytoplankton respiration and carbon stock (including phytoplankton, zooplankton, bacterial biomass and particulate organic matter).”

P1, L20: Please avoid using personal pronouns. The sentence can be written as follows “Moreover, an acidification trend (signal) was observed in the upper water column…”.

P1, L22-23: Please write it as follows “The projected changes are stronger in the Eastern Mediterranean due to the limited influence of the exchanges in the Strait of Gibraltar in that part of the basin.”

Introduction:

P1, L31: These are some key references (Lascaratos, 1993; Nittis and Lascaratos, 1998) but they are old. I would suggest to also add newer ones, i.e.

• Fedele, G., Mauri, E., Notarstefano, G., and Poulain, P. M.: Characterization of the Atlantic Water and Levantine Intermediate Water in the Mediterranean Sea using Argo Float Data, Ocean Sci. Discuss. [preprint], https://doi.org/10.5194/os-2021-68, in review, 2021
• Fach, B. A., Orek, H., Yilmaz, E., Tezcan, D., Salihoglu, I., Salihoglu, B., & Latif, M. A. (2021). Water mass variability and Levantine Intermediate Water formation in the Eastern Mediterranean between 2015 and 2017. Journal of Geophysical Research: Oceans, 126, e2020JC016472. https://doi.org/10.1029/2020JC016472
• Velaoras, D., Papadopoulos, V.P., Kontoyiannis, H., Cardin, V. and Civitarese, G., 2019. Water masses and hydrography during April and June 2016 in the cretan sea and cretan passage (Eastern Mediterranean Sea). Deep Sea Research Part II: Topical Studies in Oceanography, 164, pp.25-40.

P1-2, L41-42: These are some key references but they are old. I would suggest to also add newer ones, i.e.

• For Marine heatwaves: Ibrahim, Omneya, Bayoumy Mohamed, and Hazem Nagy. 2021. "Spatial Variability and Trends of Marine Heat Waves in the Eastern Mediterranean Sea over 39 Years" Journal of Marine Science and Engineering 9, no. 6: 643. https://doi.org/10.3390/jmse9060643
• For Med. droughts: Mathbout, Shifa, Joan A. Lopez-Bustins, Dominic Royé, and Javier Martin-Vide. 2021. "Mediterranean-Scale Drought: Regional Datasets for Exceptional Meteorological Drought Events during 1975–2019" Atmosphere 12, no. 8: 941. https://doi.org/10.3390/atmos12080941

P3, L75: Please remove “thus far”.

P3, L97: I would suggest to replace “provide” by “sustain”.

P3, L74-97: There are some missing articles in this section. For example, Howes et al. (2015) also derived the same conclusions using the RCPs 4.5 and 8.5. There is also Macias et al. (2018) who used two different global circulation models (GCMs; equivalent to RCP4.5 and RCP8.5), and other studies.

• Herrmann, M., Estournel, C., Adloff, F., and Diaz, F. (2014), Impact of climate change on the northwestern Mediterranean Sea pelagic planktonic ecosystem and associated carbon cycle, J. Geophys. Res. Oceans, 119, 5815– 5836, doi:10.1002/2014JC010016.
• Howes EL, Stemmann L, Assailly C, Irisson JO, Dima M, Bijma J, Gattuso JP (2015) Pteropod time series from the North Western Mediterranean (1967-2003): impacts of pH and climate variability. Mar Ecol Prog Ser 531:193-206. https://doi.org/10.3354/meps11322
• Macias, D., Garcia-Gorriz, E. and Stips, A. (2018), Major fertilization sources and mechanisms for Mediterranean Sea coastal ecosystems. Limnol. Oceanogr., 63: 897-914. https://doi.org/10.1002/lno.10677

P3, L99: Please write “All the above-mentioned works…” instead of “All of these previous works…”.

P4, L112: Please write “non-living” instead of “nonliving”.

Data and methods:

P6, L184: Please write “non-living” instead of “nonliving”. And edit it through the entire text.

P6, L187: Please write “physico-chemical” instead of “physical-chemical”.

P6, L191-192: Please pay attention to the subscripts throughout the manuscript, i.e. CaCO₃

P7-8, section 2.4: The subscripts are sometimes too small to read. Please rectify it.

Results:

P8, L284: It would be helpful to add the ranges, SDs, maybe in a table. What are the precisions of T and S derived from this model?

P10-11, L317-322, Fig. 2:

• I would recommend to add the Chl-a unit on the next to the bar dedicated for a & b.
• Also add the unit of the depth, on the profiles or the caption.
• You need to add in the caption that the vertical profiles are shown for the Mediterranean scale, Western Mediterranean, and Eastern Mediterranean. This was not mentioned in the corresponding text as well.
• Please write the unit appropriately for “µmol kg^-1” in the caption as well as throughout the ms.

P11, L338: Please add “such as…” instead of “as…”.

P12, L354-355: Please make it clearer, i.e. General freshening of the upper layers and saltening of the intermediate layers are observed over the entire basin during the MID-FUTURE period (Fig. S3 in the supplementary materials).

P13, L382: Please edit: “Only for the Aegean Sea, the changes in the winter mixed layer maximum depth are less marked, …”.

P14, Fig. 4: Please correct the presentation of scenarios in the plots: RCP 4.5 and 8.5 instead of RCP 45 and 85.

P14, L404: Please delete both “the” in “the nutrients at the river mouths.”

P16-17, Fig. 5-6: Please add the unit on both color bars. Please correct the presentation of scenarios: RCP 4.5 and 8.5 instead of RCP 45 and 85.

P18, L445-449: Please try to make this sentence clearer, i.e. Between 2055 and 2075, the peak in both nutrients’ concentration, for RCP4.5, timely corresponds to a peak in the inflow of nutrients at the Gibraltar strait (Fig. S7). Additionally, in both scenarios the intermediate layer of the Western basin, after 2035, experiences a negative tendency in the nutrient concentration which is greater than 0.01 mmol m^-3 for PO₄ and 0.1 mmol m^-3 NO₃, this is related to a reduced westward transport of nutrients associated with LIW (Fig.S5).

P20, L479-80: Please add references here.

P20, L484: Please correct “in vertical processes’ intensity”.

P21, L489: Do you mean “in both basins”? Please correct.

P21, Fig. 9: Please correct the presentation of scenarios: RCP 4.5 and 8.5 instead of RCP 45 and 85.

P21, L498: Please remove “the” from “both the scenarios”.

P21, L499: Please add a “,” after scenario.

P22, L508: Please add a “,” after scenarios.

P22, L514: Please add a “,” after scenario.

P22, L519: Please add a “,” after scenarios.

P23, L526, Fig.11: Please add the unit on both color bars. Please correct the presentation of scenarios: RCP 4.5 and 8.5 instead of RCP 45 and 85. Please unify the term “Mid-FUTURE” in the captions and the plots.

P23, L535: Please write it “parts of the Levantine basin”.

P23, L538: Please add a “,” after FUTURE.

P24, L562: Please keep either “of approximately” or “of about”.

P26, L580: Please add a “,” after FUTURE.

P27, Fig. 14: Why did you choose the abbreviation BACTC to bacterial biomass? It is not conventional. I would suggest to make it “BACT”.

P27, L592: Please remove the “,” after also, and add it before “the decline”.

P28, L616-617: The influence of the air-sea CO2 exchanges on DIC concentrations in the Mediterranean were already highlighted in multiple studies, i.e.

from models

• D’Ortenzio, F., Antoine, D. and Marullo, S., 2008. Satellite-driven modeling of the upper ocean mixed layer and air–sea CO2 flux in the Mediterranean Sea. Deep Sea Research Part I: Oceanographic Research Papers, 55(4), pp.405-434.

for observations

• Wimart-Rousseau, C., Lajaunie-Salla, K., Marrec, P., Wagener, T., Raimbault, P., Lagadec, V., Lafont, M., Garcia, N., Diaz, F., Pinazo, C. and Yohia, C., 2020. Temporal variability of the carbonate system and air-sea CO2 exchanges in a Mediterranean human-impacted coastal site. Estuarine, Coastal and Shelf Science, 236, p.106641.
• Hassoun, A.E.R., Fakhri, M., Abboud-Abi Saab, M., Gemayel, E. and De Carlo, E.H., 2019. The carbonate system of the Eastern-most Mediterranean Sea, Levantine Sub-basin: Variations and drivers. Deep Sea Research Part II: Topical Studies in Oceanography, 164, pp.54-73.
• De Carlo, E.H., Mousseau, L., Passafiume, O., Drupp, P.S. and Gattuso, J.P., 2013. Carbonate chemistry and air–sea CO 2 flux in a NW Mediterranean bay over a four-year period: 2007–2011. Aquatic geochemistry, 19(5), pp.399-442.

P28, L620: Please replace “fairly” by “equal”.

P28, L621: Please replace “consistently”.

P28, L625: Please remove “than in”.

P30, L649: Please remove the “s” from “produces”.

P30, L651: Please correct the subscript in “pCO₂”.

P30, L654-655: Please re-write, i.e. “consistent with the estimates of Solidoro et al. (2021).”

P30, L657-658: Do you mean “by the end of the 21^st century for RCP8.5?”. Please rectify.

P31, Fig.17: Please write the unit appropriately for “µmol kg^-1” in the caption. Please correct the presentation of scenarios: RCP 4.5 and 8.5 instead of RCP 45 and 85.

P31, L673-674: I would recommend to also check Goyet et al. (2016), as the pattern of your results are somehow in harmony.

P31, L675: I would suggest to mention “by the end of the 21^st century for RCP8.5” or “by the end of the century for RCP8.5”.

P3, L682-684: Please unify the term “Mid-FUTURE” in the captions and the plots.

Discussions and conclusions:

P33, L689: Please add a “,” after “In this study”.

P33, L693: Please add a “,” after “To the best of the authors’ knowledge”.

P33, L693-696: Please re-write this sentence taking into consideration the major comments above.

P33, L718: Please remove the “s” at the end of “shows”.

P35, L768-769: This sentence is important for coastal ecosystems, and needs thus better elaboration, and references too.

P35, L771-772: The exchanges via the Strait of Gibraltar are surely crucial, but there are other factors that should be taken into consideration such as the difference in the ventilation period between both basins, among other factors (i.e. Pujo-Pay et al., 2011; Álvarez et al., 2014; Stöven and Tanhua, 2014; Cardin et al., 2015; Hassoun et al., 2015; Goyet et al., 2016; etc.).

• Álvarez, M., Sanleón-Bartolomé, H., Tanhua, T., Mintrop, L., Luchetta, A., Cantoni, C., Schroeder, K., and Civitarese, G.: The CO2 system in the Mediterranean Sea: a basin wide perspective, Ocean Sci., 10, 69–92, https://doi.org/10.5194/os-10-69-2014, 2014.
• Cardin, V., Civitarese, G., Hainbucher, D., Bensi, M., and Rubino, A.: Thermohaline properties in the Eastern Mediterranean in the last three decades: is the basin returning to the pre-EMT situation?, Ocean Sci., 11, 53–66, https://doi.org/10.5194/os-11-53-2015, 2015.
• Goyet, C., Hassoun, A., Gemayel, E., Touratier, F., Abboud-Abi Saab, M. and Guglielmi, V., 2016. Thermodynamic forecasts of the mediterranean sea acidification. Mediterranean Marine Science, 17(2), pp.508-518.
• Hassoun, A.E.R., Gemayel, E., Krasakopoulou, E., Goyet, C., Abboud-Abi Saab, M., Guglielmi, V., Touratier, F. and Falco, C., 2015. Acidification of the Mediterranean Sea from anthropogenic carbon penetration. Deep Sea Research Part I: Oceanographic Research Papers, 102, pp.1-15.
• Pujo-Pay, M., Conan, P., Oriol, L., Cornet-Barthaux, V., Falco, C., Ghiglione, J.F., Goyet, C., Moutin, T. and Prieur, L., 2011. Integrated survey of elemental stoichiometry (C, N, P) from the western to eastern Mediterranean Sea. Biogeosciences, 8(4), pp.883-899.
• Stöven, T. and Tanhua, T.: Ventilation of the Mediterranean Sea constrained by multiple transient tracer measurements, Ocean Sci., 10, 439–457, https://doi.org/10.5194/os-10-439-2014, 2014.

P35, L778: Please also compare it with Mediterranean projections (i.e. Goyet et al., 2016).

P35, L790-792: Is it possible to estimate these uncertainties? It would be great to mention the level of overestimation derived from the model compared to the present conditions.

P35, L794-801: A recent study by Gazeau et al. (2021) is a good fit in this section as well, as it highlights the potential impact of aerosol deposition (dust in this case) both in present and future climate conditions in the Mediterranean.

• Gazeau, F., Ridame, C., Van Wambeke, F., Alliouane, S., Stolpe, C., Irisson, J.-O., Marro, S., Grisoni, J.-M., De Liège, G., Nunige, S., Djaoudi, K., Pulido-Villena, E., Dinasquet, J., Obernosterer, I., Catala, P., and Guieu, C.: Impact of dust addition on Mediterranean plankton communities under present and future conditions of pH and temperature: an experimental overview, Biogeosciences, 18, 5011–5034, https://doi.org/10.5194/bg-18-5011-2021, 2021.

P36, L826: Please write it “such as”.

Supplementary document:

Fig. S2: Please add the temperature unit on both color bars (°C). Please correct the presentation of scenarios: RCP 4.5 and 8.5 instead of RCP 45 and 85. Please correct “blue” instead of “blu”.

Fig. S3: Please correct the presentation of scenarios: RCP 4.5 and 8.5 instead of RCP 45 and 85.

Fig. S4: Please correct the presentation of scenarios: RCP 4.5 and 8.5 instead of RCP 45 and 85. Please correct “blue line” instead of “blu line”.

Fig. S8: Please add the unit on both color bars. Please correct the presentation of scenarios: RCP 4.5 and 8.5 instead of RCP 45 and 85.

Fig. S9-14: Please add the unit on both color bars. Please correct the presentation of scenarios: RCP 4.5 and 8.5 instead of RCP 45 and 85.

Fig. S14: Why are you mentioning the DIC unit in ug kg^-1 here while it is in µmol kg^-1 in the text? Please adopt the second one.

General comments: 

This paper describes new high-resolution coupled (physic and biogeochemistry) simulations made under the RCP 4.5 and 8.5 scenarios for the Mediterranean Sea (MED). This work is in line with the studies that have been carried out over the last ten years to assess the effects of climate change on the circulation and biogeochemistry of the MED. The main improvement here is the use of a higher spatial resolution than the previous studies. I acknowledge the huge amount of work and computational time that is required to run such simulations and I do think that this study is a step forward to a better understanding of the effect of climate change on the MED. However, despite the overall good quality of this work, this paper could be improve.

The use of higher resolution simulations could significantly improve our understanding of the MED in the context of climate change, but the authors don't really highlight the interest of such high resolution in the manuscript. The authors propose an analysis based on two averaged integrated depths 0-100 m and 200-600 m for 8 Mediterranean sub-basins losing in my opinion the opportunity to fully take advantage of their model resolution. Furthermore, the discussion as proposed here lacks rigorous comparison and discussion with the existing literature which could emphasize more the importance of high resolution. I also have concerns about the methodology, for example (but not only) the choice of the reference periods (2005-2020) that have been taken into account including the beginning of the scenarios, and the choice of allowing only the dissolved inorganic carbon to change at Gibraltar Strait.

 Specific comments : 

Biogeochemical spin-up:

 Some important information about the spin-up is missing for biogeochemistry. Please specify how you did the spin-up and how long it runs before being stable. It is briefly mentioned in section 3.1 but this should be presented before.

 Boundary conditions: 

 The boundary conditions are kept constant except for the DIC. Why all the variables of the carbonate cycle aren’t treated the same way? Furthermore, did you use boundary conditions from a global RCP 8.5 for the MED RCP 8.5 simulation and a global RCP 4.5 for the MED RCP 4.5 simulation? 

Overall, this needs to be discussed as it could change the DIC / ALK ratio at Gibraltar Strait that might be important for pH variations in the MED. Please, specify how the DIC concentration evolves at Gibraltar Strait. A table representing nutrients and carbonate system variables for Gibraltar Strait and rivers will be appreciated.

Periods selection PRESENT MID and FAR:  

 A PRESENT period of 15 years was chosen and compared with MID and FAR periods of 20 years. How this could impact your results? Why don't you choose 20 years as well for the PRESENT period?

 During PRESENT period (2005-2020), the simulations are already in scenarios mode, as the RCP scenarios start in 2005. Therefore, the PRESENT period encompasses change linked to climate change which could lead to bias for evaluating the effect induced by climate change. Indeed, major differences between the 2 simulations considering the 2 RCP scenarios may appear during the PRESENT period, as observed for zonal stream function on your figure 4. Usually, the reference period is chosen among the hindcast part of the simulation to avoid those issues ( Richon et al. 2019; Pages et al. 2020b, and others). Did you run a hindcast part before the scenarios? Please, at least, discuss your choice of the reference period and how it may affect your results. 

Statistical significance:

 There is no indication in the text about the significance of the differences obtained for comparison between the MID and FAR periods. The word ‘significant’ is used but without statistics. In order to evaluate whether the numeric differences are substantial, it is necessary to calculate some parameters such as the t-student (See line 517 as an example) and to indicate if the variations are significant in the text and in the figures.

 Some temporal correlations are indicated over the manuscript, for example between oxygen concentration and decrease of MLD, but any values are given (r and p-value) to assess the results. Please add some statistics. 

Model validations: 

Please add more quantitative validations with for example a Taylor diagram. 

 The model is validated based on 2005-2020 averages made with the CTRL simulation forced with RCP scenario. As mentioned before, it is not well suited to use that for model validations because the scenarios already have an impact. Could you discuss the potential implications? 

 Discussion limitations: 

 The discussion is interesting overall but there is a lack of links between the paragraphs and some parts could be more detailed. 

 Line 718, it is indicated that the signals are in agreement with the previous literature. This is not exact at least for the NPP as  Macias et al. (2015) pointed out a decrease of the primary production rates in the Western Basin and an increase in the Eastern Basin, and Pages et al., 2020 a general decrease of the NPP. Please, address the variables one by one, the differences between the models results are a strength if they are discussed. 

The use of 1/16 resolution grid is one of the strengths of this study but I think this needs to be more discussed as said in the General comments section. This point has been highlighted in the introduction and is supposed to be the main improvement of this study. However, in the current state of the discussion, it seems that the same conclusions may have been reached with a lower resolution.  

 Could you accentuate the discussion around the difference observed between the RCP 8.5 and the RCP 4.5 ?

Mixed layer depth:

 How did you define the mixed layer depth? Please add this information to the manuscript.  

 NPP :

 The effects of the model equations on the NPP trends could be more discussed. This is an important difference between the models which needs to be addressed. Your model equations assume that the planktonic community will remain the same over the next century without adaptation to warmer temperatures, which is unlikely. Furthermore, O’Connor et al., (2011) is cited to explain particulate organic matter decrease despite NPP increase. This paper focuses on terrestrial plants and herbivores which are very different from phytoplankton. Please, add other more appropriate references. 

Nutrients : 

 A nutrients peak is obtained with the RCP 4.5 simulation between 2055 and 2075 (figure S7) and described in lines 455-456. How do you explain this peak? At Gibraltar’s Strait the nutrient concentrations are supposed to be fixed, so did the surface water flux change? Could you explain why?

To the best of my knowledge, the Atlantic inflow is hypothesized to increase a bit over the next century due to stronger evaporation in the eastern basin that will increase the SSH gradient. Might this strong peak (and the even higher peak in 2095) be related to model instability?

Could you also explain why we didn't observe the peak in the RCP 8.5 simulation where the effect of climate change should be even stronger? 

This is concerning as this nutrient input affects the response of the WMED (see for example your figure 12 where a peak is visible for the organic matter). 

Vertical stratification affects the sinking velocity of particles: 

 It is suggested that vertical stratification affects the sinking velocity of particles (lines 544-545). To my knowledge, even if stratification might impact the downward flux of particles, most of the models didn’t take that into account. Could you please explain how it is taken into account in BFM?

DIC and pH :

Line 626: “Disentangle the temperature and DIC contributions …” how did you do this? 

  Figures :

There are 18 figures in the manuscript and 14 in SM witch is a lot. 

The figures like figure 5 (14 are concerned) presented in the supplementary material could be sum-up in a table. The manuscript will become easy to read (mostly the result section) with fewer references to the supplementary material. 

 In general, there are issues with the figures. The captions are generally not detailed enough and need to be re-written. The units need to be on the figure and not only in the captions. There are cosmetic issues on the figures. Here are a few random examples (more are given in the minors comments section) :

           - Figure 3 the labels are too small, units for temperature?

           - Figure 4 units? and please, add a map showing the location of the transect

           - Figure 16, there is no label for the year, and apparently, you give twice the WMED, d,e,f should be EMED

           - Figure S7, the y-axis label is too stretched, use scientific notation you will have more space. 

           - Figure S8 (S9) the size of both color bars seems to be different.    

Technical corrections :  

 Figure 1: Please highlight extend of the Atlantic buffer zone in the figure. In general, I would like to see the unit below the scale/color bar of a figure, this comment applies to all the figures of the manuscript.

Line 75: “all focusing on high emission…”  replace all by “mostly” as Macias et al., 2015 use the RCP 4.5 that is not a high emission scenario. 

Line 76: A1B climate change: please explain

Line 83: A2 emission scenario: please, relate to the RCP scenarios used here

Lines 94-95: “Moreover, the work also projected …” change by “Moreover, Solidoro et al 2020 also … ”

Line 109, line 115: same paragraph as before

Line 116: You say 70 vertical levels and later (line 150) you say 72 please correct that.

Line 118: You need to define the acronym you use MFS16 and OGSTM-BFM and add citations for both. 

Line 138: Same thing as before for OPA this time.

Line 145: Same for NEMO.

Line 153: Same for CMCC-CM. I will not continue to list those errors, please be sure that every acronym is defined as it is first used. 

Line 188: chemical reactions

Line 191-192: CaCO3

Lines 294- 276: I think this is an interesting 

approach but this paragraph is difficult 

to read. Could you try to make it a bit more clear? I think that bringing the S1 figure here could be helpful for the reader. 

Line 297: Specify the periods of the satellite climatology. 

Line 310 - 315: The spin-up information needs to be completed and added before in the manuscript.

Figure 2: Please add the units on the figure for the profiles, and the map. This remark applies to all the figures of the manuscript. 

Figure 2 Add the source of the dataset cited in the caption. 

Figure 3 : The font size of “Year” is too small, this applies to other figures that show time series. If I understand correctly, the variables that you show here are corrected from the CTRL bias ( X_anom(k)scen ), please mention it in the caption. Those comments apply to the other times series of the manuscript. 

Furthermore, why did you apply a 10-years running mean here and not in figure 7? 

This 10-years running mean is not mentioned in the manuscript (unless I miss it) so it is confusing.

Line 338: Give more citations here.  

Line 348: remove “Fig. S2 in the supplementary material” by (Fig. S2 and S3). The “in the supplementary material” is not necessary and you are not using it all the time, so, remove it everywhere in the manuscript. 

Line 360: replace “increased freshwater deficit” by “decreasing freshwater discharge” 

Figure 4: I would like the unit near the color bar. You need to show the location of this zonal stream function over a map. 

Figures 5: Again, place the units near the color bars. For the caption, there is not enough information to understand the figure. What are the values in green squares? What is the MID-FUTURE/ FAR-FUTURE ? The unit given is just for the 4 top figures. Please change the caption to take this into account. This comment is valid for all the other figures in the manuscript and in the supplementary material that looks like figure 5. 

Line 384: Define EMT.   

Line 418: Fig.S6 add space.

Line 507 : (Fig. 12) should be Fig. 11 

Line 555 : (both phyto- and zoo-) remove the – 

Line 658: Should be Figure 17  

Line 724: change Pages et al 2020 for Pages et al 2020b. This applies to all the manuscript.

Line 856: in the bibliography, certain references format are not coherent. For example at line 856 all the authors aren’t listed there is an et al. This is the same for lines 890, 902,916, 933, 1006, 1080, 1103. 

Line 902: Punctuation is wrong    

Line 982: Format issue