The salinity of the inflow in Gibraltar decreases (-9.0·10^-3 psuyr^-1) due to the freshening of the inflowing Atlantic waters. This fresher Atlantic water is the main contributor to the surface freshening in the GoL. Otherwise, the outflow waters become saltier (+2.0 ·10^-3psuyr^-1) because of the salinization of LIW and deeper waters as shown in Figure 3, thus the salt transport through the Strait to Atlantic Ocean increases (Table 1).

4 Discussion and Conclusions

The response of the NWMed deep water formation to climate change has been explored in Soto-Navarro et al. (2020). The authors found in an ensemble of six Regional Climate Models a robust dramatic decrease of DWF by the end of the 21st century. However, the mechanisms involved were not revealed. As the time evolution of the DWF is strongly model dependent, single model runs can be used to identify in a physically consistent way the causes of this collapse. Therefore, we study these mechanisms in one of the simulations used in Soto-Navarro et al. (2020). Specifically, we analyze the simulation with the regionally coupled system ROM under the RCP8.5 scenario.

Our simulation is able to reproduce the several DWF episodes during the present period: deep convection occurs every year from 2009 to 2014 (Figure 2a and Table S1), what is statistically in good agreement with observations (Houpert et al., 2016; Somot et al., 2018; Margirier et al., 2020).

Starting from the early 2040s, we find a dramatic reduction of MLD_max in the GoL (Figure 2a). The maximum MLD for the present climate (1976-2005) is 2385 ± 630 m while for 2070-2099 under the RCP8.5 scenario is 297± 47 m. That is, the MLD_maxsimulated by ROM experiences a reduction of almost 90% in the GoL. These values are in close agreement with values reported in Soto-Navarro et al. (2020).

Exploring the possible mechanisms (t. e. air-sea fluxes and ocean preconditioning) that lead to this dramatic MLD_maxdecrease we find that the air-sea fluxes do not seem to be the factor responsible for the DWF collapse, as the buoyancy loss does not change significantly (Figure 2b). Changes in the buoyancy loss also cannot explain the changes in DWF strength, as strong or weak DWF events can happen with similar BL values. In agreement with Somot et al. (2018), our results show that in any case no deep water is formed (MLD_max < 1000 m) when BL < 0.6 m²s^-2 during the 2006-2013 period.

Our results indicate that the changes in properties of the upper and intermediate water masses affecting the ocean preconditioning are key in the DWF collapse. In our simulation, the temperatures of MAW and LIW are projected to increase 2.6ºC and 2.3ºC, respectively (Figure 3a and 3b). In turn, the salinity decreases slightly (-0.01 psu) on the surface, while deeper waters become saltier (Figure 3c and 3d). As shown previously in Parras-Berrocal et al. (2020), the warming and freshening signal of MAW comes to the GoL from the Eastern Atlantic, while the increase of LIW temperature and salinity is propagated from its formation area in the Eastern Mediterranean. These changes increase the vertical density gradient between the MAW and LIW, strongly reducing the vertical mixing between these water masses. This is reflected in the stratification index for the 0-1000 m water column (Figure 3e), which from 2040s onward experiences a positive trend (0.02 m²s^-2y^-1). Our results do not show a significant change in atmospheric fluxes, changes in MAW and LIW characteristics play the main role in the DWF collapse in the NWMed. The recent study of Margirier et al. (2020) lends some support to our hypothesis: they found in glider and other platforms profiles collected over 2007-2017 that an abrupt jump of LIW temperature and salinity provoked a strong reduction of vertical mixing in the NWMed in 2014. The authors concluded that under those conditions, stronger atmospheric forcing is needed to trigger deep convection. Amitai et al. (2021) have also recently demonstrated that LIW characteristics play a key role in enabling or disabling the deep convection in the GoL. In agreement with Margirier et al. (2020) and Amitai et al. (2021), our results indicate that the projected deep water formation collapse in the 21st century is controlled by the change in LIW characteristics, but we found also that changes in MAW properties play a role. In order to assess the relative contribution of LIW and MAW to the deep water formation collapse we compare the SI calculated from spatially and temporally averaged vertical profiles in the GoL in four cases (Figure S2): (i) using the values corresponding to the pre-collapse period (2006-2041); (ii) using the values corresponding to the post-collapse period (2070-2099); and creating additionally two synthetic profiles, (iii) one containing the pre-collapse characteristics of MAW (0-200 m depth) with the post-collapse properties of deeper layers (200-1000 m depth; Figure S2g), and (iv) a second one with the post-collapse MAW and pre-collapse deeper layers. As seen in Figure S2, there is a clear increase in the post-collapse SI (see also Fig. 3e), but interestingly the SI for the (iv) situation (1.20 m²s^-2, Figure S2h) is greater than for the (iii) (0.97 m²s^-2, Figure S2g). This results suggest that in the future the change in MAW properties will play a role at least as relevant as LIW in the deep water formation collapse. This collapse of NWMed deep water formation seems to have an impact on the ventilation and on the thermohaline circulation of the Mediterranean Sea.

Both inflow and outflow water transport at the Strait of Gibraltar show a decreasing trend which is larger in the outflow, increasing the net flow (Table 1). Moreover, the incoming surface Atlantic jet will transport more heat and less salt into the Mediterranean basin, causing the hydrographic changes of surface waters (MAW) in the GoL represented in Figure 3. At the same time, the salinization of intermediate (LIW) in the Mediterranean Sea contributes to the increase of MO salinity. As a result, the MO will be warmer and saltier by the end of the 21st century. During the period with DWF episodes (2006-2041) the warming of MO is gradual and does not present noticeable changes, however after the collapse of DWF it accelerates reaching an increase rate of 0.034ºCyr^-1 (Figure S1). Then, the collapse of DWF could be one of the driving factors of the changes in characteristics of fluxes at Gibraltar Strait which reflects changes in the Mediterranean overturning circulation at large scale.

Acknowledgments and Data

I. M. Parras-Berrocal, O. Álvarez and M. Bruno were supported by the Spanish National Research Plan through project TRUCO (RTI2018-100865-BC22). R. Vázquez was supported through a doctoral grant at the University Ferrara and University of Cádiz. W. Cabos have been funded by the Spanish Ministry of Science, Innovation and Universities, the Spanish State Research Agency and the European Regional Development Fund, through grant CGL2017-89583-R. D. Sein was supported in the framework of the state assignment of the Ministry of Science and Higher Education of Russia (0128-2021-0014). The model data are available online (at https://doi.org/10.5281/zenodo.5151396).

The authors declare that they have no conflict of interest.

References

Adloff, F., Somot, S., Sevault, F., Jordà, G., Aznar, R., Déqué, M., Herrmann, M., Marcos, M., Dubois, C., Padorno, E., & Alvarez-Fanjul, E. (2015). Mediterranean Sea response to climate change in an ensemble of twenty first century scenarios. Climate Dynamics, 45, 2775–2802. https://doi.org/10.1007/s00382-015- 2507-3

Amitai. Y., Ashkenazy, Y., & Gildor, H. (2021). The Effect of the Source of Deep Water in the Eastern Mediterranean on Western Mediterranean Intermediate and Deep Water. Frontiers in Marine Science, 7:615975. https://doi.org/10.3389/fmars.2020.615975

Bethoux, J.P., & Gentili, B. (1999). Functioning of the Mediterranean Sea: past and present changes related to freshwater input and climate changes. Journal of Marine Systems, 20, 33–47. https://doi.org/10.1016/S09247963(98)00069-4

Darmaraki, S., Somot, S., Sevault, F., Nabat, P., Cabos Nar- vaez, W. D., Cavicchia, L., Djurdjevic, V., Li, L., Sannino, G., & Sein, D. V. (2019). Future evolution of Marine Heatwaves in the Mediterranean Sea, Climate Dynamics, 53, 1371–1392. https://doi.org/10.1007/s00382-019-04661-z

D’Ortenzio, F., Iudicone, D., de Boyer Montegut, C., Testor, P., Antoine, D., Marullo, S., Santoreli, R., & Madec, G. (2005). Seasonal variability of the mixed layer depth in the mediterranean sea as derived from in situ profiles. Geophysical Research Letter, 32, L12605. https://doi.org/10.1029/2005GL022463

Durrieu de Madron, X., Houpert, L., Puig, P., Sanchez-Vidal, A., Testor, P., Bosse, A., Estournel, C., Somot, S., Bourrin, F., Bouin, M. N., Beauverger, M., Beguery, L., Calafat, A., Canals, M., Cassou, C., Coppola, L., Dausse, D., D’Ortenzio, F., Font, J., Heussner, S., Kunesch, S., Lefevre, D., Goff, H. L., Martin, J., Mortier, L., Palanques, A., & Raimbault, P. (2013). Interaction of dense shelf water cascading and open-sea convection in the Northwestern Mediterranean during winter 2012. Geophysical Research Letter, 40, 1379–1385. https://doi.org/10.1002/grl.50331

Giorgetta, M. A., Jungclaus, J., Reick, C. H., Legutke, S., Bader, J., Böttinger, M., Brovkin, V., Crueger, T., Esch, M., Fieg, K., Glushak, K., Gayler, V., Haak, H., Hollweg, H.-D., Ily- ina, T., Kinne, S., Kornblueh, L., Matei, D., Mauritsen, T., Mikolajewicz, U., Mueller, W., Notz, D., Pithan, F., Raddatz, T., Rast, S., Redler, R., Roeckner, E., Schmidt, H., Schnur, R., Segschneider, J., Six, K. D., Stockhause, M., Timmreck, C., Wegner, J., Widmann, H., Wieners, K.-H., Claussen, M., Marotzke, J., & Stevens, B. (2013). Climate and carbon cycle changes from 1850 to 2100 in MPI-ESM simulations for the Coupled Model Intercomparison Project phase 5. Journal of Advances in Modelling Earth Systems, 5, 572–597, https://doi.org/10.1002/jame.20038

Giorgi, F. (2006). Climate change hot-spots. Geophysical Research Letter, 33, L08707. https://doi.org/10.1029/2006GL025734

Hagemann, S., & Dümenil-Gates, L. (1998). A parameterization of the lateral waterflow for the global scale Climate Dynamics, 14, 17–31. https://doi.org/10.1007/s003820050205

Hagemann, S. & Dümenil-Gates, L. (2001). Validation of the hydrological cycle of ECMWF and NCEP reanalysis using the MPI hydrological discharge model. Journal Geophysical Research, 106, 1503–1510. https://doi.org/10.1029/2000JD900568

Herrmann, M., Sevault, F., Beuvier, J., & Somot, S. (2010). What induced the exceptional 2005 convection event in the northwestern Mediterranean basin? Answers from a modeling study. Journal Geophysical Research: Oceans, 115(C12)051. https://doi.org/10.1029/2010JC006162

Hibler, W. D. (1979). A dynamic thermodynamic sea ice model. Journal of Physical Oceanography, 9, 815–846. https://doi.org/10.1175/1520- 0485(1979)009<0815:ADTSIM>2.0.CO;2

Houpert, L., Durrieu de Madron, X., Testor, P., Bosse, A., D’Ortenzio, F., Bouin, M. N., Dausse, D., Le Goff, H., Kunesch, S., Labaste, M., & Coppola, L. (2016). Observations of open‐ocean deep convection in the northwestern Mediterranean Sea: Seasonal and interannual variability of mixing and deep water masses for the 2007‐2013 period. Journal of Geophysical Research: Oceans, 121(11), 8139-8171. https://doi.org/10.1002/2016JC011857

IPCC. (2013). Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change (Stocker TF, Qin D, Plattner G-K, Tignor M, Allen SK, Boschung J, Nauels A, Xia Y, Bex V and Midgley PM (eds)). Cambridge University Press, Cambridge and New York, pp 1535

Jacob, D. (2001). A note to the simulation of the annual and in- terannual variability of the water budget over the Baltic Sea drainage basin. Meteorology and Atmospheric Physics, 77, 61–73. https://doi.org/10.1007/s007030170017

Jungclaus, J. H., Fischer, N., Haak, H., Lohmann, K., Marotzke, J., Matei, D., Mikolajewicz, U., Notz, D., & von Storch, J. S. (2013). Characteristics of the ocean simulations in MPIOM, the ocean component of the MPI-Earth system model. Journal of Advances in Modelling Earth Systems, 5, 422–446. https://doi.org/10.1002/jame.20023

Lascaratos, A., Williams, R., & Tragou, E. (1993). A mixed-layer study of the formation of Levantine Intermediate Water. Journal Geophysical Research, 98(C8), 14739–14749. https://doi.org/10.1029/93JC00912

Leaman, K., & Schott, F. (1991). Hydrographic structure of the convection regime in the Gulf of Lions: Winter 1987. Journal of Physical Oceanography, 21(4), 575–598.

Léger, F., Lebeaupin Brossier, C., Giordani, H., Arsouze, T., Beuvier, J., Bouin, M.-N., Bresson, E., Ducrocq, V., Fourrié, N., & Nuret, M. (2016). Dense water formation in the north- western Mediterranean area during HyMeX-SOP2 in 1/36° ocean simulations: Sensitivity to initial conditions. Journal Geophysical Research: Oceans, 121(8), 5549–5569. https://doi.org/10.1002/2015JC011542

L’Hévéder, B. Li, L. Sevault, F. & Somot, S. (2013). Interannual variability of deep convection in the Northwestern Mediterranean simulated with a coupled AORCM. Climate Dynamics, 41(3–4), 937–960. https://doi.org/10.1007/s00382-012-1527-5

Maier-Reimer, E., Kriest, I., Segschneider, J., & Wetzel, P. (2005). The HAMburg Ocean Carbon Cycle Model HAMOCC5.1 Technical Description Release 1.1, Ber. Erdsystemforschung, 14, available at: http://hdl.handle.net/11858/00-001M-0000-0011-FF5C-D

Margirier, F., Testor, P., Heslop, E., Mallil, K., Bosse, A., Houpert, L., Mortier, L., Bouin, M.-B., Coppola, L., D’Ortenzio, F., Durrie de Madron, X., Mourre, B., Prieur, L., Raimbault, P. & Taillandier, V. (2020). Abrupt warming and salinification of intermediate waters interplays with decline of deep convection in the Northwestern Mediterranean Sea. Scientific Reports 10, 20923. https://doi.org/10.1038/s41598-020-77859-5

Marshall, J., & Schott, F. (1999). Open-ocean convection: observations, theory, and models. Reviews of Geophysics, 37(1), 1–64. https://doi.org/10.1029/98RG02739

Marsland, S. J., Haak, H., Jungclaus, J. H., Latif, M., & Roeske, F. (2003). The Max-Planck- Institute global ocean/sea ice model with orthogonal curvilinear coordinates. Ocean Modelling, 5 (2), 91–127, https://doi.org/10.1016/S1463-5003(02)00015-X

MEDOC Group. (1970). Observations of formation of deep-water in the Mediterranean Sea. Nature, 227, 1037–1040. https://doi.org/10.1038/2271037a0

Menna, M. & Poulain, P. M. (2010). Mediterranean intermediate circulation estimated from Argo data in 2003–2010. Ocean Science, 6, 331–343. https://doi.org/10.5194/os-6-331-2010

Mertens, C., & Schott, F. (1998). Interannual variability of deep-water formation in the northwestern Mediterranean. Journal of Physical Oceanography, 28, 1410–1423, https://doi.org/10.1175/1520-0485(1998)028<1410:IVODWF>2.0.CO;2

Mikolajewicz, U., Sein, D. V., Jacob, D., Königk, T., Podzun, R. & Semmler, T. (2005). Simulating Arctic sea ice variability with a coupled regional atmosphere-ocean-sea ice model. Meteorologische Zeitschrift, 14 (6), 793–800. https://doi.org/10.1127/0941-2948/2005/0083

Millot, C. (1999). Circulation in the Western Mediterranean Sea, Journal of Marine Systems, 20 (1-4), 423-440. https://doi.org/10.1016/S0924-7963(98)00078-5

Millot, C. (2005). Circulation in the mediterranean sea: Evidences, debates and unanswered questions, Scientia Marirna, 69, 5–21. https://doi.org/10.3989/scimar.2005.69s15

Millot, C. (2014). Levantine Intermediate Water characteristics: an astounding general misunderstanding! (addendum), Scientia Marina, 78(2), 165–171. https://doi.org/10.3989/scimar.04045.30H

Parras-Berrocal, I. M., Vazquez, R., Cabos, W., Sein, D., Mañanes, R., Perez-Sanz, J., & Izquierdo, A. (2020). The climate change signal in the Mediterranean Sea in a regionally coupled atmosphere–ocean model, Ocean Science, 16, 743–765. https://doi.org/10.5194/os-16-743-2020

Rechid, D., & Jacob, D.: Influence of monthly varying vegetation on the simulated climate in Europe. Meteorologische Zeitschrift, 15(1), 99–116. https://doi.org/10.1127/0941-2948/2006/0091

Rhein, M. (1995). Deep water formation in western Mediterranean. Journal of Geophysical Research, 100-C4, 6943-6959. https://doi.org/10.1029/94JC03198

Robinson, A., Leslie, W., Theocharis, A., & Lascaratos, A. (2001). Encyclopedia of ocean sciences. Academic Press Ltd., London chap Mediterranean Sea Circulation

Sánchez-Gómez, E., Somot, S., Josey, S. A., Dubois, C., Elguindi, N., & Déqué, M. (2011). Evaluation of Mediterranean Sea water and heat budgets simulated by an ensemble of high resolution regional climate models. Climate Dynamics, 37, 2067–2086. https://doi.org/10.1007/s00382-011-1012-6

Schott, F., & Leaman, K. D. (1991). Observations with Moored Acoustic Doppler Current Profilers in the Convection Regime in the Golfe du Lion. Journal of Physical Oceanography, 21, 558–574, https://doi.org/10.1175/1520- 0485(1991)021<0558:OWMADC>2.0.CO;2

Sein, D. V., Mikolajewicz, U., Gröger, M., Fast, I., Cabos, W., Pinto, J. G., Hagemann, S., Semmler, T., Izquierdo, A., & Jacob, D. (2015). Regionally coupled atmosphere-ocean- sea ice-marine biogeochemistry model ROM: 1. Description and validation, Journal of Advance in Modelling Earth Systems, 7(1), 268–304. https://doi.org/10.1002/2014MS000357

Sein, D. V., Gröger, M., Cabos, W., Alvarez-Garcia, F. J., Hagemann, S., Pinto, J. G., Izquierdo, A., de la Vara, A., Koldunov, N. V., Dvornikov, A. Y., Limareva, N., Alekseeva, E., Martinez-Lopez, B. & Jacob, D. (2020). Regionally coupled atmosphere-ocean- sea ice-marine biogeochemistry model ROM: 2. Studying the Climate Change Signal in the North Atlantic and Europe, Journal of Advance in Modelling Earth Systems, 12, e2019MS001646. https://doi.org/10.1029/2019Ms001646

Seyfried, L., Marsaleix, P., Richard, E., & Estournel, C. (2017). Modelling deep-water formation in the north-west Mediterranean Sea with a new air–sea coupled model: sensitivity to turbulent flux parameterizations. Ocean Science, 13, 1093–1112. https://doi.org/10.5194/os-13-1093-2017

Shaltout, M., & Omstedt, A. (2014). Recent sea surface temperature trends and future scenarios for the Mediterranean Sea. Oceanologia, 56, 411–443. https://doi.org/10.5697/oc.56-3.411

Somot, S., Sevault, F., & Déqué, M. (2006). Transient climate change scenario simulation of the Mediterranean Sea for the 21st century using a high-resolution ocean circulation model. Climate Dynamics, 27, 851–879. https://doi.org/10.1007/s00382-006-0167-z

Somot, S., Sevault, F., Déqué, M., & Crépon, M. (2008). 21st century climate change scenario for the Mediterranean using a coupled atmosphere–ocean regional climate model. Global Planetary Change, 63, 112–126. https://doi.org/10.1016/j.gloplacha.2007.10.003

Somot, S., Houpert, L., Sevault, F., Testor, P., Bosse, A., Taupier-Letage, I., Bouin, M. N., Waldman, R., Cassou, C., Sanchez-Gomez, E., Durrieu de Madron, X., Adloff, F., Nabat, P. & Herrman, M. (2018). Characterizing, modelling and understanding the climate variability of the deep water formation in the North-Western Mediterranean Sea. Climate Dynamics, 51, 1179–1210. https://doi.org/10.1007/s00382-016-3295-0

Soto-Navarro, J., Jordá, G., Amores, A., Cabos, W., Somot, S., Sevault, F., Macias, D., Djurdjevic, V., Sannino, G., Li, L. & Sein, D. (2020). Evolution of Mediterranean Sea water properties under climate change scenarios in the Med-CORDEX ensemble. Climate Dynamics, 54, 2135–2165. https://doi.org/10.1007/s00382-019-05105-4

Taylor, K., Stouffer, R., & Meehl, G. (2012). An overview of CMIP5 and the experiments design. Bulletin of the American Meteorological Society, 93, 485–498. https://doi.org/10.1175/BAMS-D-11-00094.1

Valcke, S. (2013). The OASIS3 coupler: a European climate modelling community software. Geoscientific Model Development, 6, 373–388. https://doi.org/10.5194/gmd-6-373-2013

Turner, J. (1973). Buoyancy effects in fluids: Cambridge monographs on mechanics and applied mathematics. Cambridge University Press, Cambridge

Vargas-Yáñez, M., Zunino, P., Schroeder, K., López-Jurado, J. L., Plaza, F., Serra, M., Castro, C., García-Martínez, M. C., Moya, F. & Salat, J. (2012). Extreme Western Intermediate Water formation in winter 2010. Journal of Marines Systems, 105-10, 52-59. https://doi.org/10.1016/j.jmarsys.2012.05.010

Waldman, R., Somot, S., Herrmann, M., Bosse, A., Caniaux, G., Estournel, C., Houpert, L., Prieur, L., Sevault, F., & Testor, P. (2017). Modelling the intense 2012–2013 dense water formation event in the northwestern Mediterranean Sea: Evaluation with an ensemble a simulation approach. Journal of Geophysical Research: Oceans, 122, 1297–1324. https://doi.org/10.1002/2016JC012437

Waldman, R., Somot, S., Herrmann, M., Sevault, F., & Isachsen, P. E. (2018). On the Chaotic Variability of Deep Convection in the Mediterranean Sea. Geophysical Research Letters, 45,2433-2443. https://doi.org/10.1002/2017GL076319

Wetzel, P., Haak, H., Jungclaus, J., & Maier-Reimer, E. (2004). The Max-Planck-institute global ocean/sea ice model. Model MPI-OM Technical report. http://www.mpimet.mpg.de/fileadmin/models/ MPIOM/DRAFT_MPIOM_TECHNICAL_REPORT.pdf