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

AMOC hysteresis in an eddy-permitting GCM and monitoring indicators

Laura Jackson and Richard Wood
Laura Jackson and Richard Wood
  • Hadley Centre, Met Office, Exeter, UK

We conduct idealised experiments with HadGEM3-GC2, which is a pre-CMIP6 eddy-permitting GCM, to test for the presence of thresholds in the AMOC. We add fresh water to the North Atlantic for different rates and lengths of time, and then examine the AMOC recovery. In some cases the AMOC recovers to its original strength, however if the AMOC weakens sufficiently it does not recover and stays in a weak state for up to 300 years.

We test various indictors that have been proposed for monitoring the AMOC with this ensemble of experiments (and other scenarios). In particular we ask whether fingerprints can provide early warning or faster detection of weakening or recovery, or indications of crossing the threshold. We find metrics that perform best are the temperature metrics based on large scale differences, the large scale meridional density gradient, and the vertical density difference in the Labrador Sea. Mixed layer depth is also useful for indicating whether the AMOC recovers after weakening. 

How to cite: Jackson, L. and Wood, R.: AMOC hysteresis in an eddy-permitting GCM and monitoring indicators, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-5498,, 2020


Display file

Comments on the display

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

displays version 1 – uploaded on 27 Apr 2020
  • CC1: Comment on EGU2020-5498, David Smeed, 01 May 2020

    I was pleased to see  your presentation, and particularly intersted in the monitoring indicators.   On slides  14/15 I was not sure what all the indicators were (e.g. M26, pintg, m_dipole).  Could you define them?

    • AC1: Reply to CC1, Laura Jackson, 01 May 2020

      Thanks David and sorry not to include that information! M26 is the AMOC at 26N, pintg in the meridional difference in depth integrated pressure (ie like Butler et al), m_dipole is a meridional dipole in density. The paper is hopefully very nearly accepted so can send you a copy if you're interested.


  • CC2: Comment on EGU2020-5498, Didier Swingedouw, 01 May 2020

    Hi Laura,

    I like your idea of a new MIP with idealized and realisitic hosing experiments. I believe this is now very important to progress on our understanding of the spread among models concerning AMOC fate, and this type of experiments might be very useful to evaluate strength of feedbacks of the AMOC system as well as the non-linearity and resilience capacity you mention here.

    Thus, I will be very happy to contribute to such a MIP using IPSL-CM6 model. You can count on me.



    • AC2: Reply to CC2, Laura Jackson, 04 May 2020

      Great thanks Didier! You were on my list of people to contact. I was planning on sending out something about this a couple of month ago, but then everything went crazy. I'm hoping to at least get started on this this year so will be in touch

  • AC4: Qn: Why is the threshold 8Sv? What about the salinity feedback?, Laura Jackson, 05 May 2020

    The value of 8Sv is, I'm sure, model dependent. From analysing the various feedbacks it seems that the salinity advection by the overturning in the subtropics is the most important. The AMOC weakening influences the salinity/frsehwater advection and when this is weakened enough that the salinity import by the AMOC is smaller than the other feedbacks the AMOC stays weak.

  • AC6: Have you compared to CMIP6-like model?, Laura Jackson, 05 May 2020

    This model (HadGEM2-GC2) is very similar to the eddy-permitting one we have submitted to CMIP6 (HadGEM3-GC3.1). I'm hoping to run some other experiments with the newer model and also hope to find volunteers to run a couple of experiments with their models for comparison!