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

The recent AMOC variability in the Subpolar Gyre: results across the OVIDE section

Pascale Lherminier1, Herlé Mercier2, Fiz F. Perez3, and Marcos Fontela4
Pascale Lherminier et al.
  • 3IIM, CSIC, Vigo, Portugal
  • 4CCMAR, Universidade do Algarve, Faro, Portugal

According to the subpolar AMOC index built from ARGO and altimetry, the AMOC amplitude across the OVIDE section (from Greenland to Portugal) was similar to that of the mid-1990s between 2014 and 2017, i.e. 4-5 Sv above the level of the 2000s. It then returned to average values in 2018. The same index computed independently from the biennial summer cruises over 2002-2018 confirms this statement. Interestingly, despite the concomitant cold and fresh anomaly in the subpolar Atlantic, the heat flux across OVIDE remains correlated with the AMOC amplitude. This can be explained by the paths taken by the North Atlantic Current and the transport anomalies in the subarctic front. In 2014, the OVIDE section was complemented by a section from Greenland to Newfoundland (GA01), showing how the water of the lower limb of the AMOC was densified by deep convection in the Labrador Sea. The spatial patterns of volume, heat, salt and oxygen transport anomalies after 2014 will be discussed at the light of the 2000s average.

How to cite: Lherminier, P., Mercier, H., Perez, F. F., and Fontela, M.: The recent AMOC variability in the Subpolar Gyre: results across the OVIDE section, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-19663,, 2020

Display materials

Display file

Comments on the display material

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

Display material version 1 – uploaded on 04 May 2020
  • CC1: Comment on EGU2020-19663, Christopher W. Hughes, 05 May 2020

    Hi Pascale - how does the overturning vary in time when the vertical shear is constant? Is this a measure of the barotropic currents moving laterally to be in regions if different depth, while the density field is held constant?

    Cheers,   Chris Hughes

    • AC1: Reply to CC1, Pascale Lherminier, 05 May 2020

      Hi Chris, I'm not sure I understand your question, and it would be a pleasure to discuss with you. Some elements concerning what I understand in your question:

      - the AMOC is calculated in sigma coordinates, since precisely the barotropic circulation blurs the signal in z-coordinates.

      - we assume that the transport by the barotropic circulation is mainly isopycnal, although it is questionnable during the restratification period in early spring.


      • CC4: Reply to AC1, Christopher W. Hughes, 05 May 2020

        The blue line in your figure uses surface currents from AVISO, but a climatology for the hydrography (I'm assuming you mean an annually repeating series of 12 monthly values?). So, with a climatology, the annual mean thermal wind is constant, but the current at the surface is varying. That combination means that all the variation is in a mode which is independent of depth (what I called barotropic), superimposed on a steady flow with constant vertical shear.

        In more tropical regions, this would be a poor approximation - you'd expect the bottom current to be relatively small, and the surface flow to be mainly a result of thermal wind integrated up from the bottom. Assuming barotropic variability would then over-represent the variability below the thermocline, and give a very wrong MOC, in either z or sigma coordinates.

        So I guess what I'm getting at is that you're saying you can get a decent estimate of AMOC variability by assuming the variability in currents is decoupled from the variability in density, so it becomes a question of how barotropic currents project onto the different climatological density classes.

        In effect, the stratification is fixed in time but depends on all spatial coordinates sigma = sigma(x,z), and the currents are independent of depth v=v(x,t). This is completely complementary to the subtropical case where it's possible to assume density is dominated by a function of z only sigma=sigma(z), so it's only the zonally-averaged currents at each depth that matter, so you'd get the same answer if you assumed v=v(z,t).

        Have I understood that correctly? Then the barotropic currents cause overturning changes because the sigma(z) function where the flow is to the north is different from sigma(z) where the returning, barotropic southward flow is? 

        • AC4: Reply to CC4, Pascale Lherminier, 05 May 2020

          Dear Chris,

          yes, you understood it right. This index works because
          (1) the vertical shear is quite small in the subpolar gyre where the lower branch returns to the south (between 100m and the bottom in the Irminger Sea because the densities are high there),
          (2) the NAC that transports the upper branch at light densities is in a different part of the section (south east) where the vertical shear is quite strong but it does not really matter since the southward transport in the subpolar gyre is much stronger than the error we make in the lower part of the NAC by adding the surface velocity to a "constant" velocity profile.

          Note that we only have the density fields in the first 2000m (ARGO limitation) so we integrate from the surface to get the AMOC amplitude.  

          Thanks a lot for your feedback: it helps me to understand what I shall strenghen in the future paper. I'm actually working to make that point clearer.

          And you are right: this method cannot work in subtropical areas with a "monthly" climatology, for the reasons you wrote and also because the upper and lower branches are nearly on top of each other in an area poorly covered by ARGO floats. I'm trying to apply it in the Labrador Sea but (1) I have a problem with ARGO data in the western boundary, and (2) the maximum of the integrated transport is much deeper although it should stay above 2000m if I rely on the data I have.

  • CC2: OVIDE vs 45N, Eleanor Frajka-Williams, 05 May 2020

    How does this compare to the Desbruyeres et al. 2019 ( estimate for 45N overturning?  The two time series appear to show opposing low frequency variations. 

    • AC2: Reply to CC2, Pascale Lherminier, 05 May 2020

      Hi Eleanor,

      this is actually an excellent question, and we will interact with Damien on this to build a full comparison in the next paper.

  • CC3: Comment on EGU2020-19663, Eric de Boisseson, 05 May 2020

    Bonjour Pascale,

    Have you also made comparisons with ocean reanalysis products? To see how they fare against observation-based estimates of the AMOC. Reanalyses tend to focus on comparisons with RAPID data. I think your data is as valuable. It would be good to encourage such comparisons.

    A bientot,


    • AC3: Reply to CC3, Pascale Lherminier, 05 May 2020

      Dear Eric,

      noce to hear from you. I hope you are doing well.

      TO answer your question: unfortunately not. Data takes up all my time. But you are absolutly right !!

      Take care

      • CC5: Reply to AC3, Laura Jackson, 05 May 2020

        In case you do want to look at reanalyses this paper may be of use We looked at the AMOC at 26N, 50N and OSNAP but haven't looked at OVIDE