EGU2020-7896
https://doi.org/10.5194/egusphere-egu2020-7896
EGU General Assembly 2020
© Author(s) 2020. This work is distributed under
the Creative Commons Attribution 4.0 License.

Transient Response of Atlantic Heat and Freshwater Transports in Future Climate Scenarios

Jennifer Mecking1,2 and Sybren Drijfhout2
Jennifer Mecking and Sybren Drijfhout
  • 1National Oceanography Center, Marine Systems Modelling, Southampton, United Kingdom of Great Britain and Northern Ireland (jennifer.mecking@noc.ac.uk)
  • 2University of Southampton, Ocean and Earth Science, UK

Ocean heat and freshwater transports play an important role in today’s climate system.  The Atlantic meridional heat transport transports 1.2 PW of heat northward leading to the warm climate we experience in Europe today, while the freshwater transport due to the Atlantic Meridional Overturning Circulation (AMOC) is often used as an indicator for the stability of the AMOC.  Future climate projections show that the AMOC is expected to weaken over the next several decades.  These changes to the AMOC as well as other circulations changes will not only impact the heat and freshwater transports in the Atlantic but also the temperature and salinity structure.  Using both CMIP5 and CMIP6 data this study untangles the impacts of velocity changes versus temperature/ salinity in future climate projections on Atlantic heat and freshwater transports.  Initial results show that changes in velocity dominate heat transport changes while the changes in salinity structure play a large role in freshwater transports with the impact of velocity changes being latitude and model dependent.

How to cite: Mecking, J. and Drijfhout, S.: Transient Response of Atlantic Heat and Freshwater Transports in Future Climate Scenarios, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-7896, https://doi.org/10.5194/egusphere-egu2020-7896, 2020

Display materials

Display file

Comments on the display material

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

Display material version 2 – uploaded on 04 May 2020
Same as version 1 but now with significance indicated on slide 8
  • AC1: Reply to Laura Jackson's CC2, Jennifer Mecking, 04 May 2020

    Thanks Laura, I have now added significance to slide 8 in presentation version 2

  • AC2: Reply to André Jüling's CC1, Jennifer Mecking, 04 May 2020

    Hi André, Thanks for your interest in my work.  It would be nice to see a comparison of similar models at different resolutions to see how the transports compare, so it would be interesting to see what comes out of your work.  Here are my replies to your questions:

    1. Have you found significant improvements between CMIP5 and 6 affecting the salinity bias that you investigate in your 2017 paper (e.g. improved surface freshwater fluxes)? I am asking in particular because your mean CMIP6 total freshwater transports do not change in reponse to forcing, but the CMIP5 reduce.

    From my analysis up to this point I didn't notice an improvement in salinity bias between CMIP5 and 6, in fact the bias seems to be strengthend in CMIP6, especially the strong salty bias centered around a depth of 1000 m and the subtropical North Atlantic.  However, the surface fresh bias in the very South Atlantic appears weaker in CMIP6.  These details will have to be investigated in more detail.

    2. The CMIP5&6 models all have 1° oceans, do you have thoughts on how eddying oceans would influence the results?

    Yes the CMIP5&6 models are all about 1° for the ocean, which isn't very good for eddies...  I think that eddies in higher resolution models will reduce the salinity bias seen in the subtropical North Atlantic at a depth of about 1000 m.  It is at these latitudes that the additional freshwater transports due to eddies converge (see Tréguier et al. 2012 in JGR-Oceans and Mecking et al. 2016 in Climate Dynamics) and therefore at these latitudes I would expect the biases to reduce.  I would also expect the Mov  in the subtropical North Atlantic to be even more negative, closer to the observational estimate as the water will be fresher.  However, I have not put much thought towards how this might impact the response to future climate senarios.

    3. The Mov change in the North Atlantic subtropical gyre in CMIP6 is very different from the CMIP5 change, and largely offset by the Maz change. Do you know why this is?

    I don't know.  However, this is a region here we have a large influence from the change in AMOC strength and this decrease in fresh water transport may be changing the structure of the salinity in a way that is not zonally symetrical.  Definately something to look into more.

    4. I am trying to wrap my head around the meaning of the correlations (and why the total  freshwater would be consistently positive in CMIP5 and negative in CMIP6). The negative correlations in the Northern Hemisphere surprised me at first, but I suppose they are negative because the freshwater transport is negative there. What did you correlate exactly, annual data?
    Thanks again for presenting your work and considering my questions.

    See the new version of the presentation.  The correlations in the total fresh water transport are mostly not significant at the 1% level so I would not read too much into them.  I calculated the correlations by computing the difference between 2070-2099 in the future climate change senarios and 1970-1999 for the historical simulations for the AMOC at 26N for each model and also the same for the change in each transport type at each latitude for each model separately and then correlated these to things to come up with a value at each latitude.

Display material version 1 – uploaded on 04 May 2020
  • CC1: Comment on EGU2020-7896, André Jüling, 04 May 2020

    Hi Jenny,
    thanks for sharing your work, very interesting! I have a few questions regarding the freshwater transport terms (which I am also looking at in an eddying version of CESM1).
    1. Have you found significant improvements between CMIP5 and 6 affecting the salinity bias that you investigate in your 2017 paper (e.g. improved surface freshwater fluxes)? I am asking in particular because your mean CMIP6 total freshwater transports do not change in reponse to forcing, but the CMIP5 reduce.
    2. The CMIP5&6 models all have 1° oceans, do you have thoughts on how eddying oceans would influence the results?
    3. The Mov change in the North Atlantic subtropical gyre in CMIP6 is very different from the CMIP5 change, and largely offset by the Maz change. Do you know why this is?
    4. I am trying to wrap my head around the meaning of the correlations (and why the total  freshwater would be consistently positive in CMIP5 and negative in CMIP6). The negative correlations in the Northern Hemisphere surprised me at first, but I suppose they are negative because the freshwater transport is negative there. What did you correlate exactly, annual data?
    Thanks again for presenting your work and considering my questions.
    André

  • CC2: Comment on EGU2020-7896, Laura Jackson, 04 May 2020

    Hi Jenny

    Would be useful to include a measure of significance for the correlations. Interesting work though!