Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Sill-controlled salinity contrasts followed post-Messinian flooding of the Mediterranean


A mile-high marine cascade terminated the Messinian salinity crisis 5.33 Myr ago, due to partial collapse of the Gibraltar sill that had isolated a largely desiccated Mediterranean from the Atlantic Ocean. Atlantic waters may have refilled the basin within 2 years. Prevailing hypotheses suggest that normal marine conditions were established across the Mediterranean immediately after the catastrophic flooding. Here we use proxy data and fluid physics-based modelling to show that normal conditions were likely for the western Mediterranean, but that flooding caused a massive transfer of salt from the western to the eastern Mediterranean across the Sicily sill, which became a hyper-salinity-stratified basin. Hyper-stratification inhibited deep-water ventilation, causing anomalously long-lasting organic-rich (sapropel) sediment deposition. Model data agreement indicates that hyper-stratification breakdown by diapycnal diffusion required 26,000 years. An alternative hypothesis that Atlantic reconnection occurred after the Mediterranean had largely been refilled is inconsistent with our observations, as this would have led to hyper-stratification and sapropel formation in both basins. Our findings offer insight into the role of stratification in delaying the re-establishment of normal marine conditions following abrupt refilling of a previously desiccated ocean basin.

This is a preview of subscription content, access via your institution

Access options

Buy article

Get time limited or full article access on ReadCube.


All prices are NET prices.

Fig. 1: XRF core-scanning and magnetic data across the M/P boundary from ODP Site 967.
Fig. 2: Evolution of Mediterranean sub-basins during the Zanclean flood.
Fig. 3: Mediterranean evolution during the evolving phase.

Data availability

ODP Site 967 data from this study are available from Panagea ( under ‘Scanning XRF and environmental magnetic data across the Miocene-Pliocene boundary from ODP Site 967 (eastern Mediterranean)’ and are also available as online Supplementary Data accompanying this article. Source data are provided with this paper.

Code availability

All the figures in this manuscript are reproducible via Jupyter notebooks and instructions provided in the Github repository Med_evolution_megaflood61 (


  1. Krijgsman, W., Hilgen, F. J., Raffi, I., Sierro, F. J. & Wilson, D. S. Chronology, causes and progression of the Messinian salinity crisis. Nature 400, 652–655 (1999).

    Article  Google Scholar 

  2. Manzi, V. et al. Age refinement of the Messinian salinity crisis onset in the Mediterranean. Terra Nova 25, 315–322 (2013).

    Article  Google Scholar 

  3. Garcia-Castellanos, D. et al. Catastrophic flood of the Mediterranean after the Messinian salinity crisis. Nature 462, 778–781 (2009).

    Article  Google Scholar 

  4. Hsü, K. J., Ryan, W. B. F. & Cita, M. B. Late Miocene desiccation of the Mediterranean. Nature 242, 240–244 (1973).

    Article  Google Scholar 

  5. Urgeles, R. et al. New constraints on the Messinian sealevel drawdown from 3D seismic data of the Ebro margin, western Mediterranean. Basin Res. 23, 123–145 (2011).

    Article  Google Scholar 

  6. Barber, P. M. Messinian subaerial erosion of the proto-Nile delta. Mar. Geol. 44, 253–272 (1981).

    Article  Google Scholar 

  7. Clauzon, G. Le canyon messinien du Rhone: une preuve decisive du ‘desiccated deep-basin model’ (Hsu, Cita et Ryan 1973). Bull. Soc. Geol. Fr. (1982).

  8. Gargani, J. & Rigollet, C. Mediterranean Sea level variations during the Messinian salinity crisis. Geophys. Res. Lett. 34, L10405 (2007).

    Article  Google Scholar 

  9. Blanc, P. L. Improved modelling of the Messinian salinity crisis and conceptual implications. Palaeogeogr. Palaeoclimatol. Palaeoecol. 238, 349–372 (2006).

    Article  Google Scholar 

  10. Meijer, P. T. & Krijgsman, W. A quantitative analysis of the desiccation and re-filling of the Mediterranean during the Messinian salinity crisis. Earth Planet. Sci. Lett. 240, 510–520 (2005).

    Article  Google Scholar 

  11. Blanc, P. L. The opening of the Plio-Quaternary Gibraltar Strait: assessing the size of a cataclysm. Geodin. Acta 15, 303–317 (2002).

    Article  Google Scholar 

  12. García-Veigas, J., Cendón, D. I., Gibert, L., Lowenstein, T. K. & Artiaga, D. Geochemical indicators in western Mediterranean Messinian evaporites: implications for the salinity crisis. Mar. Geol. 403, 197–214 (2018).

    Article  Google Scholar 

  13. Marzocchi, A., Flecker, R., van Baak, C. G. C., Lunt, D. J. & Krijgsman, W. Mediterranean outflow pump: an alternative mechanism for the Lago-Mare and the end of the Messinian salinity crisis. Geology 44, 523–526 (2016).

    Article  Google Scholar 

  14. Andreetto, F. et al. Freshening of the Mediterranean salt giant: controversies and certainties around the terminal (Upper Gypsum and Lago-Mare) phases of the Messinian salinity crisis. Earth-Sci. Rev. 216, 103577 (2021).

    Article  Google Scholar 

  15. Micallef, A. et al. Evidence of the Zanclean megaflood in the eastern Mediterranean basin. Sci. Rep. 8, 1078 (2018).

    Article  Google Scholar 

  16. Garcia-Castellanos, D. et al. The Zanclean megaflood of the Mediterranean – searching for independent evidence. Earth-Sci. Rev. 201, 103061 (2020).

    Article  Google Scholar 

  17. Camerlenghi, A. et al. Seismic markers of the Messinian salinity crisis in the deep Ionian basin. Basin Res. 32, 716–738 (2020).

    Article  Google Scholar 

  18. Pierre, C., Rouchy, J. M. & Blanc-Valleron, M. M. Sedimentological and stable isotope changes at the Messinian/Pliocene boundary in the eastern Mediterranean (holes 968A, 969A, and 969B). Proc. Ocean Drill. Progr. Sci. Results 160, 3–8 (1998).

    Google Scholar 

  19. Iaccarino, S. M. & Bossio, A. Paleoenvironment of uppermost Messinian sequences in the western Mediterranean (site 974, 975, and 978). Proc. Ocean Drill. Progr. Sci. Results 161, 529–541 (1999).

    Google Scholar 

  20. Hsu, K. et al. Messinian Paleoenvironments, initial reports. Deep Sea Drill. Proj. 42-1, 1003–1035 (1978);

  21. Emeis, K., Roberston, A. H. & Richter, C. Site 967, initial reports. Proc. Ocean Drill. Progr. 160, 215–287 (1996).

    Google Scholar 

  22. Spezzaferri, S., Cita, M. B. & McKenzie, J. A. The Miocene/Pliocene boundary in the eastern Mediterranean: results from Sites 967 and 969. Proc. Ocean Drill. Progr. Sci. Results 160, 9–28 (1998).

    Google Scholar 

  23. Emeis, K. C., Sakamoto, T., Wehausen, R. & Brumsack, H. J. The sapropel record of the eastern Mediterranean Sea - results of Ocean Drilling Program Leg 160. Palaeogeogr. Palaeoclimatol. Palaeoecol. 158, 371–395 (2000).

    Article  Google Scholar 

  24. Grant, K. M. et al. Organic carbon burial in Mediterranean sapropels intensified during Green Sahara periods since 3.2 Myr ago. Commun. Earth Environ. 3, 11 (2022).

    Article  Google Scholar 

  25. Van Santvoort, P. J. M., De Lange, G. J., Langereis, C. G., Dekkers, M. J. & Paterne, M. Geochemical and paleomagnetic evidence for the occurrence of ‘missing’ sapropels in eastern Mediterranean sediments. Paleoceanography 12, 773–786 (1997).

    Article  Google Scholar 

  26. Larrasoaña, J. C., Roberts, A. P., Hayes, A., Wehausen, R. & Rohling, E. J. Detecting missing beats in the Mediterranean climate rhythm from magnetic identification of oxidized sapropels (Ocean Drilling Program Leg 160). Phys. Earth Planet. Inter. 156, 283–293 (2006).

    Article  Google Scholar 

  27. De Lange, G. J. et al. Synchronous basin-wide formation and redox-controlled preservation of a Mediterranean sapropel. Nat. Geosci. 1, 606–610 (2008).

    Article  Google Scholar 

  28. Mercone, D., Thomson, J., Abu-Zied, R. H., Croudace, I. W. & Rohling, E. J. High-resolution geochemical and micropalaeontological profiling of the most recent eastern Mediterranean sapropel. Mar. Geol. 177, 25–44 (2001).

    Article  Google Scholar 

  29. Rohling, E. J., Marino, G. & Grant, K. M. Mediterranean climate and oceanography, and the periodic development of anoxic events (sapropels). Earth-Sci. Rev. 143, 62–97 (2015).

    Article  Google Scholar 

  30. Larrasoaña, J. C., Roberts, A. P. & Rohling, E. J. Dynamics of Green Sahara periods and their role in hominin evolution. PLoS ONE 8, e76514 (2013).

    Article  Google Scholar 

  31. Wehausen, R. & Brumsack, H. J. Chemical cycles in Pliocene sapropel-bearing and sapropel-barren eastern Mediterranean sediments. Palaeogeogr. Palaeoclimatol. Palaeoecol. 158, 325–352 (2000).

    Article  Google Scholar 

  32. Ryan, W. B. F. Modeling the magnitude and timing of evaporative drawdown during the Messinian salinity crisis. Stratigraphy 5, 227–243 (2008).

    Google Scholar 

  33. Lourens, L. J., Wehausen, R. & Brumsack, H. J. Geological constraints on tidal dissipation and dynamical ellipticity of the Earth over the past three million years. Nature 409, 1029–1033 (2001).

    Article  Google Scholar 

  34. Hilgen, F. J. Astronomical calibration of Gauss to Matuyama sapropels in the Mediterranean and implication for the Geomagnetic Polarity Time Scale. Earth Planet. Sci. Lett. 104, 226–244 (1991).

    Article  Google Scholar 

  35. Lourens, L. J. et al. Evaluation of the Plio-Pleistocene astronomical timescale. Paleoceanography 11, 391–413 (1996).

    Article  Google Scholar 

  36. Rossignol-Strick, M., Nesteroff, W., Olive, P. & Vergnaud Grazzini, C. After the deluge: Mediterranean stagnation and sapropel formation. Nature 295, 105–110 (1982).

    Article  Google Scholar 

  37. Rossignol-Strick, M. African monsoons, an immediate climate response to orbital insolation. Nature 304, 46–49 (1983).

    Article  Google Scholar 

  38. Osborne, A. H. et al. A humid corridor across the Sahara for the migration of early modern humans out of Africa 120,000 years ago. Proc. Natl Acad. Sci. USA 105, 16444–16447 (2008).

    Article  Google Scholar 

  39. Amies, J. D., Rohling, E. J., Grant, K. M., Rodríguez-Sanz, L. & Marino, G. Quantification of African monsoon runoff during last interglacial sapropel S5. Paleoceanogr. Paleoclimatol. 34, 1487–1516 (2019).

    Article  Google Scholar 

  40. Rohling, E. J. et al. Reconstructing past planktic foraminiferal habitats using stable isotope data: a case history for Mediterranean sapropel S5. Mar. Micropaleontol. 50, 89–123 (2004).

    Article  Google Scholar 

  41. Rohling, E. J. & Gieskes, W. W. C. Late Quaternary changes in Mediterranean intermediate water density and formation rate. Paleoceanography 4, 531–545 (1989).

    Article  Google Scholar 

  42. Rohling, E. J. Shoaling of the eastern Mediterranean pycnocline due to reduction of excess evaporation: implications for sapropel formation. Paleoceanography 6, 747–753 (1991).

    Article  Google Scholar 

  43. Larrasoaña, J. C. et al. Source-to-sink magnetic properties of NE Saharan dust in eastern Mediterranean marine sediments: review and paleoenvironmental implications. Front. Earth Sci. 3, 19 (2015).

    Google Scholar 

  44. Rouchy, J. M. & Caruso, A. The Messinian salinity crisis in the Mediterranean basin: a reassessment of the data and an integrated scenario. Sediment. Geol. 188189, 35–67 (2006).

    Article  Google Scholar 

  45. Roveri, M. et al. The Messinian salinity crisis: past and future of a great challenge for marine sciences. Mar. Geol. 352, 25–58 (2014).

    Article  Google Scholar 

  46. Hayakawa, Y. S. & Matsukura, Y. Factors influencing the recession rate of Niagara Falls since the 19th century. Geomorphology 110, 212–216 (2009).

    Article  Google Scholar 

  47. Ferrari, R. & Wunsch, C. Ocean circulation kinetic energy: reservoirs, sources, and sinks. Annu. Rev. Fluid Mech. 41, 253–282 (2009).

    Article  Google Scholar 

  48. Waterhouse, A. F. et al. Global patterns of diapycnal mixing from measurements of the turbulent dissipation rate. J. Phys. Oceanogr. 44, 1854–1872 (2014).

    Article  Google Scholar 

  49. Thomson, J. et al. Redistribution and geochemical behaviour of redox-sensitive elements around S1, the most recent eastern Mediterranean sapropel. Geochim. Cosmochim. Acta 59, 3487–3501 (1995).

    Article  Google Scholar 

  50. Reitz, A., Thomson, J., Lange, G. J., De & Hensen, C. Source and development of large manganese enrichments above eastern Mediterranean sapropel S1. Paleoceanography 21, PA3007 (2006).

    Article  Google Scholar 

  51. Weltje, G. J. et al. in Micro-XRF Studies of Sediment Cores (eds. Croudace, I. W. & Rothwell, R. G.) 507–534 (Springer, 2015).

  52. Rohling, E. J., Schiebel, R. & Siddall, M. Controls on Messinian lower evaporite cycles in the Mediterranean. Earth Planet. Sci. Lett. 275, 165–171 (2008).

    Article  Google Scholar 

  53. Cushman-Roisin, B. & Beckers, J.-M. Introduction to Geophysical Fluid Dynamics: Physical and Numerical Aspects (Academic Press, 2011).

  54. Garcia-Castellanos, D. & O’Connor, J. E. Outburst floods provide erodability estimates consistent with long-term landscape evolution. Sci. Rep. 8, 10573 (2018).

    Article  Google Scholar 

  55. Peltier, W. R. & Caulfield, C. P. Mixing efficiency in stratified shear flows. Annu. Rev. Fluid Mech. 35, 135–167 (2003).

    Article  Google Scholar 

  56. Osborne, T. Estimates of the local rate of vertical diffusion from dissipation measurements. J. Phys. Oceanogr. 10, 83–89 (1980).

    Article  Google Scholar 

  57. MacKinnon, J. A. et al. Climate process team on internal wave-driven ocean mixing. Bull. Am. Meteorol. Soc. 98, 2429–2454 (2017).

    Article  Google Scholar 

  58. Grothe, A. et al. Paratethys pacing of the Messinian salinity crisis: low salinity waters contributing to gypsum precipitation? Earth Planet. Sci. Lett. 532, 116029 (2020).

    Article  Google Scholar 

  59. Garcia-Castellanos, D. & Villaseñor, A. Messinian salinity crisis regulated by competing tectonics and erosion at the Gibraltar arc. Nature 480, 359–363 (2011).

    Article  Google Scholar 

  60. Duermeijer, C. E. & Langereis, C. G. Astronomical dating of a tectonic rotation on Sicily and consequences for the timing and extent of a middle Pliocene deformation phase. Tectonophysics 298, 243–258 (1998).

    Article  Google Scholar 

  61. Amarathunga, U. & Hogg, A. M. Resurgence of Mediterranean after the Zanclean megaflood - Jupyter notebooks (Python). Zenodo (2022).

  62. Grant, K. M. et al. A 3 million year index for North African humidity/aridity and the implication of potential pan-African humid periods. Quat. Sci. Rev. 171, 100–118 (2017).

    Article  Google Scholar 

  63. Laskar, J., Fienga, A., Gastineau, M. & Manche, H. La2010: a new orbital solution for the long-term motion of the Earth. Astron. Astrophys. 532, A89 (2011).

Download references


We thank D. Garcia-Castellanos for providing essential instructions on developing the Mediterranean reflooding model. We also thank P. Meijer for providing the latest Mediterranean hypsometry reconstruction. Study material on fluid dynamics provided by B. Cushman-Roisin encouraged initial model development. This work contributes to Australian Research Council projects FL120100050 and DP2000101157 (E.J.R.), DE190100042 (K.M.G.), DP190100874 (A.P.R.), and the Australia‐New Zealand IODP Consortium (ANZIC) Legacy/Special Analytical Funding grant LE160100067 (Katharine Mary Grant (K.M.G.), Laura Rodriguez Sanz (L.R.S.)).

Author information

Authors and Affiliations



U.A. designed and led the study and wrote the paper; U.A. developed the hypothesis and performed the modelling, with guidance from A.M.H. and E.J.R.; E.J.R., A.M.H., A.P.R., K.M.G. and D.H. contributed to data interpretation; K.M.G. calibrated scanning XRF data; S.G. performed WD-XRF analyses; X.Z. and P.H. assisted with magnetic measurements; D.L. performed XRF core scanning; D.L. and T.W. developed the ODP967 composite depth splice; all authors contributed to manuscript development.

Corresponding author

Correspondence to Udara Amarathunga.

Ethics declarations

Competing interests

The authors declare no competing interests.

Peer review

Peer review information

Nature Geoscience thanks Angelo Camerlenghi, Rachel Flecker and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Primary Handling Editor: James Super, in collaboration with the Nature Geoscience team.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Extended data

Extended Data Fig. 1 Locations of DSDP and ODP sites with complete records across the Miocene-Pliocene transition.

Illustration of the Mediterranean basin with locations of ODP and DSDP sites where complete records across the Miocene/Pliocene boundary have been retrieved (see Supplementary Table 1). XRF core scanning data for this study were obtained from ODP Site 967 (Eratosthenes Seamount northern flanks). The cross-section was drawn using Ocean Data View 5.3.0.

Extended Data Fig. 2 Sapropel record for ODP Site 967.

a, Ba record for ODP967 from scanning XRF data for the last 3 million years62. b, Sapropel stratigraphy for ODP Site 967. From 5.3 to 3.2 Ma23, sediments are barren of sapropels (black = sapropels, red = red intervals, blue = ghost sapropels). c, Stack of core images for ODP967 arranged according to composite depth, with sapropel occurrence. Sapropel-barren interval is indicated by the section shaded in pink. d, Enlarged view of the sapropel-barren interval. Location of the ‘mystery sapropel’ is indicated by blue shading.

Extended Data Fig. 3 Chronology.

Three peaks in the Ba/Ti record were tuned to insolation maxima (precession minima, June 21, 65°N)63 between 5140 and 5182 ka. Below this (splice depth 125.03–127.23 m), clear Ba/Ti peaks are absent, so ages were interpolated linearly assuming a sedimentation rate equivalent to that for 5162–5182 ka. This assumption is validated by the resulting insolation maximum alignment at 5312 ka (splice depth 127.23 m) with the next Ba/Ti peak. The succeeding Ba peak (at splice depth 128.35 m) is the largest, so the maximum of the smoothed curve was tuned to 5333 ka based on the timing of the Mediterranean reflooding event3. Available ODP 967 biostratigraphic datums22 (a, b, c, d) validate our chronology (see Supplementary Table 2 for details).

Source data

Extended Data Fig. 4 Conceptual model for Mediterranean refilling.

a, Sketch of main Zanclean flooding stages (ZFS). ZFS 1 ends as the wMed level reaches the Sicily Sill. ZFS 2 ends as the eMed level reaches Sicily Sill. In ZFS 3, both basins rise to the Atlantic level concurrently. MRB, Messinian Residual Brines; wMed, Western Mediterranean; eMed, Eastern Mediterranean. b, Sketch of basin evolution during the flooding phase. As flood waters flow into the basin, mixing with residual brines (purple) occurs. The mixing extent depends on the kinetic energy of flow entering the brine surface. For increased kinetic energy, a dz depth of brines will be added to the mixed layer (green). The mixed layer thickness increases accordingly. KE(t + 1), kinetic energy of flow entering the brine; KE(t), kinetic energy of flow in the previous timestep; Z(t)wMed, wMed level at a given time; Z(t)eMed, eMed level at a given time; Zmix(t)wMed, wMed mixed layer extent at a given time; Z(t)wBrine, wMed brine thickness at a given time; Z(t)eBrine, eMed brine thickness at a given time; and ZAtlantic, Atlantic level.

Extended Data Fig. 5 Salinity profiles for wMed and eMed at the end of the Flooding Phase for different ME values.

For our main model, we use ME = 20% (0.2), which is the widely accepted value for shear-driven mixing in stratified fluids55,56. Here we test the effect of ME change on basin evolution, for the 10–30 % ME range. a, Salinity profiles (wMed) at the end of the Flooding Phase for different ME values. At 10% ME, a ~700-m-thick residual brine layer remains in the bottom of the wMed. The salinity profile above the pycnocline is enlarged in this case. For 20% and 30% ME, the wMed does not contain residual brine. b, Salinity profiles (eMed) at the end of the Flooding Phase for different ME values. For each ME value, strong stratification occurs at the Sicily Sill level, where dense fluids lie toward the bottom (salinity profile colours: blue-green = wMed mixed layer; orange = eMed mixed layer; purple = residual Messinian brines; during the final flood stage, mixing occurs only in the wMed, from which mixed waters overflow toward the eMed above Sicily Sill – indicated by an upper blue-green eMed layer).

Extended Data Fig. 6 Summary of sensitivity test (Part 1) results.

a, wMed salinity profile (black) where the basin is filled to the Sicily Sill level. The thin surface layer is composed of evaporated Paratethyan waters. Blue line is wMed hypsometry. b, Evolution of the wMed level (black), mixing depth at different mixing efficiencies (0.2, 0.3 and 1 ME), velocity and flow kinetic energy with time, for the basin configuration in a. Even at 100% mixing efficiency (1 ME), flow energy fails to erode the deep brine layer. c, Comparison of refilling abruptness between a scenario where the basin is filled to the Sicily Sill level (thick lines), and a deeply desiccated basin before the catastrophic termination (dotted lines). The length of the flooding mode increases to 6400 days for a filled basin, while discharge and flow velocity are much less compared to a catastrophic termination. The shaded section demarcates the interval (dotted lines) used to compute basin evolution in the main text.

Extended Data Fig. 7 Sensitivity test Part 1 and Part 2 results.

(a,b; Sensitivity test Part 1 - Testing for different initial salinity profiles) a, Initial wMed salinity profiles used for the test; surface layer (430–560 m) and residual brine layer (below 1750 m) salinities were kept unchanged at 40 and 140 PSU, respectively. Mid-layer salinity was changed, as shown in the Figure (ISP 1 to ISP 6, ISP; initial salinity profile). For the mid-layer, salinity was increased linearly from 40 PSU at 560 m, to 50 PSU (ISP 1), 60 PSU (ISP 2), 80 PSU (ISP 3), 100 PSU (ISP 4), and 120 PSU (ISP 5), at the deep brine surface (1750 m). ISP 6 is the same as used in Extended Data Fig. 6. The black dotted line is wMed hypsometry. b, Evolution of the mixing depth at 100% mixing efficiency for different starting salinity profiles in a. For ISP 1, mixing depth reaches the brine surface, but the flood energy is insufficient to erode the brine layer. Maximum mixing depth does not extend below 1500 m for any other starting salinity profile, implying that the flood energy is insufficient to reach deep wMed brines below 1750 m. The same colours for ISPs are used for mixing depth evolution lines. The black dotted line indicates the wMed level rise as Atlantic waters fill the Mediterranean. (c,d; Sensitivity test Part 2) c, Approximation of the final eMed salinity below the Sicily Sill at the end of basin refilling (Zanclean flooding), as a function of initial Mediterranean residual fluid salinity. Residual fluid salinity in wMed and eMed are considered equal. Below 140 PSU, all wMed salt will be transferred to the eMed. d, eMed evolving phase when initial residual fluid salinity is set below 140 PSU (140 PSU was used for the main text model calculations - see Methods for reasoning). Duration of salt removal is tested at initial salinities of 60, 80, 100, and 120 PSU. The dotted line represents the maximum eMed surface salinity. The purple box above the time axis represents the expected salt removal duration for salinities between 60 and 120 PSU.

Extended Data Fig. 8 Summary of sensitivity test (Part 3) results.

a, Post-flood density profiles for the wMed and eMed for initial salinities > 140 PSU. Salinities between 160 and 240 PSU were chosen. The vertical dotted line (red) represents the winter density of wMed surface waters. Up to 170 PSU, the post-flood wMed density profile has lower than winter density values. This will result in rapid convective mixing and removal of remaining wMed salt. Above 180 PSU, greater than winter surface densities appear in the wMed. Above 220 PSU, flood energy is no longer sufficient to transfer deep wMed salt to the eMed. Strong eMed stratification is present in all cases (profile shading: blue-green, wMed mixed layer; orange, eMed mixed layer; purple, residual Messinian brines; AL, Atlantic level; SL, Sicily Sill level). b, Duration of wMed and eMed salt removal, where stratification exists. The Figure is separated into two parts due to the wide salinity range. For the wMed, model outputs indicate salt removal durations of ~4,000 years at an initial 180 PSU salinity, which can increase up to 12,000 years at 240 PSU. For the eMed, a range between ~21,500 to 27,500 years is expected between 160 and 240 PSU. The purple box above the time axis represents the expected salt removal duration (eqSSwin, equal surface salinity of winter surface water; eqSSinf, equal salinity of inflow water – see Methods for explanation).

Extended Data Fig. 9 Summary of sensitivity test (Part 4) results.

a, Flow energy in the wMed and the eMed for a ~30% shallower (300 m) or deeper (560 m) than present Sicily sill. Thick and dotted lines correspond to energy evolution for shallow and deep sill depths, respectively. b, Sea-level and mixing depth evolution for the wMed and the eMed when the Sicily sill is shallower than present (300 m). c, Sea-level and mixing depth evolution for the wMed and the eMed when the Sicily sill is deeper than present (560 m). In b and c, red curves in each panel represent the rise of basin level. Green (for wMed) and orange (for eMed) curves indicate the mixing depth evolution with time, for different mixing efficiencies (ME). Dots in coloured background, 0.1 ME; coloured-only background, 0.2 ME; Dots in white background, 0.3 ME. For a, b and c, refer to the grey colour panels at the top of the figure for Zanclean flooding stages (ZFS 1–3) for shallower and deeper sill settings. d, Density profiles at the end of the flooding phase for shallow and deep sill settings. Vertical red-dotted line represents the present wMed surface density (Mediterranean Atlantic water density in the wMed29). Green shading, wMed mixed layer; orange shading, eMed mixed layer; AL, Atlantic level; SL, Sicily sill level.

Extended Data Fig. 10 Summary of sensitivity test (Part 5) results.

a, Density profiles for the wMed at the flooding phase termination, for different starting base levels of the wMed and the eMed (eMed base level is 250 m lower than the mentioned wMed base level for each test). Vertical dotted line represents the winter wMed surface density. Beyond this line, excess salt remains in the wMed as a result of incomplete salt transfer to the eMed. b, Duration of post-flood salt diffusion for starting base levels which results in excess salt in the wMed (1,250 and 1,350 m wMed base levels below Atlantic level; correspond to 1,500 and 1,600 eMed levels, respectively). Coloured panels above the time axis represent the expected duration of salt removal from wMed to the Atlantic, for corresponding curves with same colour (eqSSwin, equal surface salinity of winter surface water; eqSSinf, equal salinity of inflow water – see Methods for explanation). SL; Sicily sill level.

Supplementary information

Supplementary Information

Supplementary Fig. 1 and Supplementary Tables 1–2.

Source data

Source Data Fig. 1

XRF and environmental magnetic data across M/P boundary from ODP Site 967.

Source Data Extended Data Fig. 3

XRF data used to develop the chronology.

Rights and permissions

Springer Nature or its licensor holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Amarathunga, U., Hogg, A.M., Rohling, E.J. et al. Sill-controlled salinity contrasts followed post-Messinian flooding of the Mediterranean. Nat. Geosci. 15, 720–725 (2022).

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI:

This article is cited by


Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing