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.

Distinct sources of interannual subtropical and subpolar Atlantic overturning variability


The Atlantic meridional overturning circulation (AMOC) is pivotal for regional and global climate due to its key role in the uptake and redistribution of heat and carbon. Establishing the causes of historical variability in AMOC strength on different timescales can tell us how the circulation may respond to natural and anthropogenic changes at the ocean surface. However, understanding observed AMOC variability is challenging because the circulation is influenced by multiple factors that co-vary and whose overlapping impacts persist for years. Here we reconstruct and unambiguously attribute intermonthly and interannual AMOC variability at two observational arrays to the recent history of surface wind stress, temperature and salinity. We use a state-of-the-art technique that computes space- and time-varying sensitivity patterns of the AMOC strength with respect to multiple surface properties from a numerical ocean circulation model constrained by observations. While, on interannual timescales, AMOC variability at 26° N is overwhelmingly dominated by a linear response to local wind stress, overturning variability at subpolar latitudes is generated by the combined effects of wind stress and surface buoyancy anomalies. Our analysis provides a quantitative attribution of subpolar AMOC variability to temperature, salinity and wind anomalies at the ocean surface.

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

Access options

Rent or buy this article

Prices vary by article type



Prices may be subject to local taxes which are calculated during checkout

Fig. 1: Schematic of the large-scale ocean circulation in the subtropical and subpolar North Atlantic.
Fig. 2: Reconstruction of overturning in the North Atlantic.
Fig. 3: Contributions of SSS and SST to variability in overturning.
Fig. 4: Spatial origins of variability in overturning at the RAPID–MOCHA and OSNAP-EAST arrays.

Similar content being viewed by others

Data availability

The OSNAP data products are publicly available at The derived data including the OSNAP-EAST overturning are furthermore available in Duke Digital Repository, The RAPID–MOCHA overturning time series is available at

Code availability

The code for the MITgcm and the scripts for post-processing model output are available at The ECCO state estimate model configuration can be downloaded from, with initial and boundary conditions available at The TAF algorithmic differentiation software is proprietary and provided by FastOpt. Code used to process data and produce figures is available from the corresponding author upon reasonable request.


  1. Lozier, M. S. Overturning in the North Atlantic. Annu. Rev. Mar. Sci. 4, 291–315 (2012).

    Article  Google Scholar 

  2. Stolpe, M. B., Medhaug, I., Sedláček, J. & Knutti, R. Multidecadal variability in global surface temperatures related to the Atlantic meridional overturning circulation. J. Clim. (2018).

  3. Cunningham, S. A. et al. Temporal variability of the Atlantic meridional overturning circulation at 26.5° N. Science 317, 935–938 (2007).

    Article  Google Scholar 

  4. Smeed, D. A. et al. The North Atlantic Ocean is in a state of reduced overturning. Geophys. Res. Lett. 45, 1527–1533 (2018).

    Article  Google Scholar 

  5. Lozier, M. S. et al. Overturning in the Subpolar North Atlantic Program: a new international ocean observing system. Bull. Am. Meteorol. Soc. 98, 737–752 (2017).

    Article  Google Scholar 

  6. Biastoch, A., Böning, C. W., Getzlaff, J., Molines, J.-M. & Madec, G. Causes of interannual–decadal variability in the meridional overturning circulation of the North Atlantic Ocean. J. Clim. 21, 6599–66615 (2008).

    Article  Google Scholar 

  7. Polo, I., Robson, J., Sutton, R. & Balmaseda, M. A. The importance of wind and buoyancy forcing for the boundary density variations and the geostrophic component of the AMOC at 26° N. J. Phys. Oceanogr. 44, 2387–2408 (2014).

    Article  Google Scholar 

  8. Ortega, P. et al. Mechanisms of decadal variability in the Labrador Sea and the wider North Atlantic in a high-resolution climate model. Clim. Dyn. 49, 2625–2647 (2017).

    Article  Google Scholar 

  9. Errico, R. M. What is an adjoint model? Bull. Am. Meteorol. Soc. 78, 2577–2591 (1997).

    Article  Google Scholar 

  10. Forget, G. et al. ECCO version 4: an integrated framework for non-linear inverse modeling and global ocean state estimation. Geosci. Model Dev. 8, 3071–3104 (2015).

    Article  Google Scholar 

  11. Pillar, H., Heimbach, P., Johnson, H. & Marshall, D. Dynamical attribution of recent variability in Atlantic overturning. J. Clim. 29, 3339–3352 (2016).

    Article  Google Scholar 

  12. Smith, T. & Heimbach, P. Atmospheric origins of variability in the South Atlantic meridional overturning circulation. J. Clim. 32, 1483–1500 (2019).

    Article  Google Scholar 

  13. Giering, R. Transformation of Algorithms in Fortran Version 1.15 (TAF Version 1.9.70) (FastOpt, 2010).

  14. Fukumori, I. et al. ECCO Version 4 Release 3 (NASA-JPL, 2017).

  15. Marshall, J., Hill, C., Perelman, L. & Adcroft, A. Hydrostatic, quasi-hydrostatic, and nonhydrostatic ocean modeling. J. Geophys. Res. 102, 5733–5752 (1997).

    Article  Google Scholar 

  16. Forget, G., Ferreira, D. & Liang, X. On the observability of turbulent transport rates by Argo: supporting evidence from an inversion experiment. Ocean Sci. 11, 839–853 (2015).

    Article  Google Scholar 

  17. Fukumori, I., Heimbach, P., Ponte, R. M. & Wunsch, C. A dynamically consistent, multivariable ocean climatology. Bull. Am. Meteorol. Soc. 99, 2107–2128 (2018).

    Article  Google Scholar 

  18. Jackson, L. C. et al. The mean state and variability of the North Atlantic circulation: a perspective from ocean reanalyses. J. Geophys. Res. Oceans 124, 9141–9170 (2019).

    Article  Google Scholar 

  19. Lozier, M. S. et al. A sea change in our view of overturning in the subpolar North Atlantic. Science 363, 516–521 (2019).

    Article  Google Scholar 

  20. Li, F., Lozier, M. S. & Johns, W. Calculating the meridional volume, heat and freshwater transports from an observing system in the subpolar North Atlantic: observing system simulation experiment. J. Atmos. Ocean. Technol. (2017).

  21. Desbruyères, D. G., Mercier, H., Maze, G. & Daniault, N. Surface predictor of overturning circulation and heat content change in the subpolar North Atlantic. Ocean Sci. 15, 809–817 (2019).

    Article  Google Scholar 

  22. Marshall, J. & Plumb, A. Atmosphere, Ocean, and Climate Dynamics: An Introductory Text (Elsevier, 2008).

    Google Scholar 

  23. Menary, M. B., Hermanson, L. & Dunstone, N. J. The impact of Labrador Sea temperature and salinity variability on density and the subpolar AMOC in a decadal prediction system. Geophys. Res. Lett. 43, 12217–12227 (2016).

    Article  Google Scholar 

  24. Zhao, J. & Johns, W. Wind‐forced interannual variability of the Atlantic meridional overturning circulation at 26.5° N. J. Geophys. Res. Oceans 119, 2403–2419 (2014).

    Article  Google Scholar 

  25. Kanzow, T. et al. Seasonal variability of the Atlantic meridional overturning circulation at 26.5° N. J. Clim. 23, 5678–5698 (2010).

    Article  Google Scholar 

  26. Jones, D. C. et al. Local and remote influences on the heat content of the Labrador sea: an adjoint sensitivity study. J. Geophys. Res. Oceans 123, 2646–2667 (2018).

    Article  Google Scholar 

  27. Loose, N., Heimbach, P., Pillar, H. R. & Nisancioglu, K. H. Quantifying dynamical proxy potential through shared adjustment physics in the North Atlantic. J. Geophys. Res. Oceans 125, e2020JC016112 (2020).

    Article  Google Scholar 

  28. Fukumori, I., Wang, O., Llovel, W., Fenty, I. & Forget, G. A near-uniform fluctuation of ocean bottom pressure and sea level across the deep ocean basins of the Arctic Ocean and the Nordic Seas. Progr. Oceanogr. 134, 152–172 (2015).

    Article  Google Scholar 

  29. Eden, C. & Willebrand, J. Mechanism of interannual to decadal variability of the North Atlantic circulation. J. Clim. 14, 2266–2280 (2001).

    Article  Google Scholar 

  30. Getzlaff, J., Böning, C. W., Eden, C. & Biastoch, A. Signal propagation related to the North Atlantic overturning. Geophys. Res. Lett. 32, L09602 (2005).

    Article  Google Scholar 

  31. MacGilchrist, G. A. et al. Locations and mechanisms of ocean ventilation in the high-latitude North Atlantic in an eddy-permitting ocean model. J. Clim. (2020).

  32. Zantopp, R., Fischer, J., Visbeck, M. & Karstensen, J. From interannual to decadal: 17 years of boundary current transports at the exit of the Labrador Sea. J. Geophys. Res. Oceans 122, 1724–1748 (2017).

    Article  Google Scholar 

  33. Zou, S., Lozier, M. S. & Buckley, M. How is meridional coherence maintained in the lower limb of the Atlantic meridional overturning circulation? Geophys. Res. Lett. 46, 244–252 (2019).

    Article  Google Scholar 

  34. Roemmich, D. et al. On the future of Argo: a global, full-depth, multi-disciplinary array. Front. Mar. Sci. 6, 439 (2019).

    Article  Google Scholar 

  35. McCarthy, G. D. et al. Measuring the Atlantic meridional overturning circulation at 26° N. Prog. Oceanogr. 130, 91–111 (2015).

    Article  Google Scholar 

  36. Kostov, Y., Johnson, H. L. & Marshall, D. P. AMOC sensitivity to surface buoyancy fluxes: the role of air–sea feedback mechanisms. Clim. Dyn. 53, 4521–4537 (2019).

    Article  Google Scholar 

  37. Weaver, A. T. & Courtier, P. Correlation modelling on the sphere using a generalized diffusion equation. Q. J. R. Meteorol. Soc. 127, 1815–1846 (2001).

    Article  Google Scholar 

  38. Heuzé, C. North Atlantic deep water formation and AMOC in CMIP5 models. Ocean Sci. 13, 609–622 (2017).

    Article  Google Scholar 

  39. Forget, G., Mercier, H. & Ferron, B. Combining Argo profiles with a general circulation model in the North Atlantic. Part 2: realistic transports and improved hydrography, between spring 2002 and spring 2003. Ocean Model. 20, 17–34 (2008).

    Article  Google Scholar 

  40. Locarnini, R. A. et al. in World Ocean Atlas 2009: Temperature. NOAA Atlas NESDIS 68 Vol. 1 (ed. Levitus, S.) 184 (US Government Printing Office 2010).

  41. Dee, D. P. et al. The ERA‐Interim reanalysis: configuration and performance of the data assimilation system. Q. J. R. Meteorol. Soc. 137, 553–597 (2011).

    Article  Google Scholar 

  42. Lund, I. A. A Monte Carlo method for testing the statistical significance of a regression equation. J. Appl. Meteor. 9, 330–332 (1970).

    Article  Google Scholar 

  43. Cornish, S. B., Kostov, Y., Johnson, H. L. & Lique, C. Response of Arctic freshwater to the Arctic oscillation in coupled climate models. J. Clim. 33, 2533–2555 (2020).

    Article  Google Scholar 

Download references


This study used the ARCHER UK National Supercomputing Service ( In our analysis, we apply the TAF software provided by FastOpt. The maps used in the figures and supplementary material were produced using the freely available software M_Map: A mapping package for MATLAB, provided by R. Pawlowicz. We thank the groups that maintain the OSNAP and RAPID–MOCHA observational networks and the developers of the ECCO version 4 state estimate. Y.K. was funded by the OSNAP project through NERC grant NE/K010948/1 and the TICTOC project through NERC grant NE/P019064/1. H.L.J. and D.P.M. were also funded by NERC grant NE/K010948/1. G.F. acknowledges support from NASA award no. 6937342 and the Simons Foundation award no. 549931. P.H., H.R.P. and T.S. were supported in part by NOAA grant NOAA/NA130AR4310135, NSF grant NSF-OCE-1924546 and a JPL/Caltech subcontract. T.S. received additional funding from an Oden Institute CSEM fellowship. N.P.H. was funded by the OSNAP NERC grant NE/K010875/1. M.S.L. and F.L. were supported by NSF grants OCE-1948335 and OCE-1924456.

Author information

Authors and Affiliations



All authors discussed the results and contributed to the preparation of the final manuscript. Y.K. took the lead in writing the text while holding regular discussions with H.L.J., D.P.M., T.S. and H.R.P. Y.K. planned, designed and performed the adjoint sensitivity analysis with the MITgcm. P.H. and G.F. developed and maintained the ECCO version 4 state estimate and the associated tools for post-processing MITgcm output on an irregular grid. T.S. adapted the MITgcm diagnostic package. N.P.H., F.L. and M.S.L. developed and applied the data analysis methodology for OSNAP observations, and F.L. provided the OSNAP-EAST overturning time series.

Corresponding author

Correspondence to Yavor Kostov.

Ethics declarations

Competing interests

The authors declare no competing interests.

Additional information

Peer review information Nature Geoscience thanks the anonymous reviewers for their contribution to the peer review of this work. Primary Handling Editor: James Super.

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

Extended data

Extended Data Fig. 1 Reconstruction skill for OSNAP-EAST observations.

Comparison between observations (yellow envelope showing ±1 standard deviation of the observational uncertainty) and our two reconstructions (outer gray contours) of OSNAP-EAST overturning [Sv] based on two different sets of sensitivity patterns: one set from objective functions in 2001–2002, and a second set from objective functions in 2006–2007. The reconstructions are interpolated onto the same 30-day windows as the observations. We consider both the mean of our two reconstructions (middle gray contour) and the spread between them (outer gray contours). Note that our reconstruction estimate uses the ECCOv4r3 mean seasonal cycle, since the observational record at OSNAP-EAST is short.

Extended Data Fig. 2 Ekman transport contribution to overturning variability at RAPID-MOCHA in ECCO.

ECCO-based comparison between variability in RAPID-MOCHA overturning (purple) and Ekman transport variability at 26°N (orange) over the time-period of the linear reconstructions in the main text. Anomalies are shown relative to the long-term mean.

Extended Data Fig. 3 Geostrophic component of overturning in the North Atlantic.

Overturning variability (purple, volume transport in Sv) at OSNAP-EAST (a) and RAPID-MOCHA (b) in the ECCO state estimate contrasted against variability in the geostrophic component of overturning (orange). The comparison in a spans the time-period of the linear reconstructions in the main text. Anomalies are shown relative to the long-term mean.

Extended Data Fig. 4 Sensitivity of the OSNAP-EAST overturning to surface heat fluxes.

Sensitivity of the OSNAP-EAST overturning in February 2007 to net surface heat fluxes [Sv per (W m-2 sustained over 1 hour)] at a lead time of nine years. Red shading indicates that heat flux into the ocean contributes to a delayed strengthening of the OSNAP-EAST overturning 9 years later. Blue shading indicates that cooling the ocean surface at that lead time causes a lagged strengthening of the OSNAP-EAST overturning. Notice the pattern tracking the Gulf Stream – North Atlantic Current advective pathway from the Caribbean to the subpolar latitudes. This long memory of past sea surface fluxes motivates the use of AMOC sensitivity to SST and SSS instead.

Extended Data Fig. 5 North Atlantic mixed layer depth and spatial origins of buoyancy-driven variability in RAPID-MOCHA overturning.

a, Climatological March mixed layer depth [m] in ECCO; b–e, Spatial sources of variability in the RAPID-MOCHA AMOC overturning: root-mean-square contribution per unit area [Sv m-2] to the convolutions in equation(1) of the main text over the period 1992–2015 using sensitivity patterns based on (b,c) 2006–2007 and (d,e) 2001–2002 AMOC objective functions. Contributions due to SST (b,d), and SSS (c,e) all relative to the seasonal cycle. The scale in all panels is linear.

Supplementary information

Supplementary Information

Supplementary Figs. 1 and 2 and references

Rights and permissions

Reprints and Permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Kostov, Y., Johnson, H.L., Marshall, D.P. et al. Distinct sources of interannual subtropical and subpolar Atlantic overturning variability. Nat. Geosci. 14, 491–495 (2021).

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