EGU General Assembly 2020
© Author(s) 2020. This work is distributed under
the Creative Commons Attribution 4.0 License.

Mechanisms for AMOC decline in the late 20th Century

Alex Megann, Adam Blaker, Simon Josey, Adrian New, and Bablu Sinha
Alex Megann et al.
  • National Oceanography Centre, Marine Systems Modelling, Southampton, United Kingdom of Great Britain and Northern Ireland (

The recent decline in the Atlantic meridional overturning circulation (AMOC) has attracted more than a little interest. The strongest AMOC recorded by the RAPID campaign at 26°N was at the start (2004/5), after which it declined about 3 Sv with a pronounced minimum in 2010. Proxies based on temperature and surface elevation have been used to extrapolate the AMOC strength before the RAPID era, and point reasonably reliably to a maximum strength in the mid 1990s, followed by a rise to a maximum at the start of the RAPID campaign in around 2005. Further back, less robust proxy data suggest that the AMOC gradually rose from the 1970s to the peak in 1990. This raises two questions: firstly, what drove these decadal variations in the overturning circulation (and hence of the ocean heat transport); and secondly whether there are observations that lead to useful predictive skill for changes in the AMOC. The surface-forced streamfunction, estimated from modelled/observed buoyancy fluxes, has been shown to be a reasonably good predictor of decadal changes in the overturning strength, preceding the latter with a lead time of about 5 years. although the reliability of the correlations before 2000 is limited by data sparsity, and especially so in the pre-satellite era.

To verify a causal link between surface forcing and decadal variations in the AMOC over longer timescales, numerical simulations present a powerful tool. A set of hindcast integrations of a global 0.25° NEMO ocean configuration has been carried out from 1958 until nearly the present day, with a selection of standard surface forcing datasets (CORE2, DFS5.2 and JRA55). These show an evolution of the AMOC strength from 1970 onwards which is consistent, both between forcing datasets and with that inferred from observations. The surface-forced streamfunction is evaluated for these experiments and is used to relate the time evolution of the AMOC to changes in the individual components of the buoyancy flux, and the surface heat loss from the Labrador and Irminger Seas is found to be the dominant predictor of AMOC changes.

How to cite: Megann, A., Blaker, A., Josey, S., New, A., and Sinha, B.: Mechanisms for AMOC decline in the late 20th Century, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-1384,, 2019

Display materials

Display file

Comments on the display material

AC: Author Comment | CC: Community Comment | Report abuse

Display material version 2 – uploaded on 01 May 2020
Updated copyright statement
  • CC1: Comment on EGU2020-1384, Laura Jackson, 01 May 2020

    Thanks Alex, that's interesting and fits with some analysis I've done on our models. My question is whether there's a spinup/repeat cycle performed with the forced model experiments to remove drift? Thanks

    • AC1: Reply to CC1, Alex Megann, 04 May 2020

      Hi Laura, thanks for your comment.

      No, there was no extended spinup - all the runs were started from rest in 1958. I used the post-1975 period for time averaging and analysis partly because of the data sparsity in earlier years, and partly because that was around the time when the time evolution of the AMOC in the three runs had roughly converged to one another.

      You mention repeat cycles. I did the 1/4° UK OMIP integration, which as you will know is integrated through five passes of CORE2 forcing. This was an almost identical configuration to the runs I describe here and to the ones we used for GO6/GO7 evaluation (Storkey et al, 2018). Comparing the AMOC between the OMIP 1/4° and 1° resolutions was interesting: at 26°N, there were differences consistent with those presented in the GO6/GO7 paper: at 1/4° the AMOC is about 6Sv stronger than at 1°. At 45°N, though, the AMOC strength and time evolution at the two resolutions is almost identical for all five forcing cycles, confirming that surface forcing, rather than dynamics, dominates the overturning strength at he higher latitude, reinforcing what we show in this presentation.    

  • CC2: Comment on EGU2020-1384, Feili Li, 05 May 2020

    Hi Alex, is Tsurf (slide#4) derived from the buoyancy (or heat) flux over the Irminger Sea alone (keeping climatology elsewhere)?  Otherwise, there might be potentially cumulative effects from the surface forcing over the Labrador Sea on your Tsurf2.

    • AC2: Reply to CC2, Alex Megann, 05 May 2020

      Hi Feili,

      Thanks for your comment. No, at the moment the surface-forced streamfunction is calculated as an integral across the whole Atlantic basin so, as you say, it will include contributions from both Labrador and Irminger basins. In the compressed version of my presentation that I uploaded, perhaps I didn't make it clear that there are really two main results here: firstly, that accumulated surface buoyancy fluxes tightly constrain the overturning north of about 40°N; and secondly that the Irminger heat loss dominates the decadal AMOC variability. An informative future step would, indeed, be to evaluate a surface-forced streamfunction integrated only across given longitude bands.


      • AC3: Reply to AC2, Bablu Sinha, 05 May 2020

        Yes that is a nice suggestion Feili

  • CC3: Comment on EGU2020-1384, Didier Swingedouw, 05 May 2020

    Dear alex,

    These are very interesting results! I have two questions:

    1) Is the evolution of the mixed layer depth in the different seas in agreement with your Tsuf index?

    (if so, this could mean that the mean stratification is also key for your Tsurf, meaning that heat fluxes are not the sole the driver, but a prerequisite to deep water formation)

    2) What is the correlation between Tsurf2 and Tover?

    (I assume it is high. If so, you might consider to compute correlation ofTsuf2 in the different seas with Tsurf2 global or Tover, which might provide explained variance (squared correlation)).

    Just in case you had no space to show this, I would be interested in such diagnostics that you might have already looked at.

    Best wishes,

    Didier Swingedouw

  • AC4: Comment on EGU2020-1384, Alex Megann, 05 May 2020

    Thanks for your interesting questions, Didier.

    1) I have only looked briefly at the MLD, but it would be informative to calculate an index for this and correlate with the overturning indices.

    2) A quantitative correlation between the two indices is the next step.

    As I replied to an earlier question, it would in principle be possible to separate different longitude bands in the calculation of the Tover indices, but I haven't thought deeply how to do this yet.

    As far as other diagnostics are concerned, we are writing a paper!


Display material version 1 – uploaded on 01 May 2020, no comments