Skip to Content

"Upper North Atlantic mixed layer already 0.5 C colder and fresher than in any winter since the inception of Argo"

A surprise return of deep convection to the subpolar North Atlantic: or never  underestimate natural climate variability. Open-ocean convection in the subpolar North Atlantic Ocean affects the meridional overturning circulation and oceanic heat flux, as well as atmospheric carbon sequestration. Recent climate warming however is thought to shallow or even switch off the convection, until winter of 2007-2008 that is, report Vage et al. in the January issue of Nature Geoscience.  Martin Visbeck (IFM-GEOMAR) and Igor Yashayaev (Bedford Institute of Oceanography) offer a point of view on the recent article. 

“Early in summer 2008, as my group was preparing to reset our moored observatories in the Irminger and Labrador Seas we received an emergency message from one of our moorings that it was adrift. This prompted us to get in touch with the Subpolar Atlantic Research community and ask for assistance. Fortunately the Canadian Research Icebreaker was in the area and within a week was able to recover the valuable data of boundary current velocities and temperatures in the North Atlantic Deep Water layers. Almost in passing Igor Yashayaev from BIO mentioned that he has just analyzed the recent ARGO (profiling float) data and saw to our surprise a return of deep convection to about 2000m. A more in depth analyses of all the ARGO data is summarized in a figure showing the temperature and salinity evolution in the two most active convection regions of the Subpolar North Atlantic (figure right).”

Not even a year after Martin Visbeck’s observations, Vage et al. publish a paper in Nature Geoscience entitled: “Surprising return of deep convection to the subpolar North Atlantic Ocean in winter 2007-2008”.

“The article provides a pretty extensive analysis of mixed layer conditions, surface heat fluxes, sea-ice and storm conditions for the region.” says Martin Visbeck.

Should we have been surprised? ”The absence of deep convective mixing over the last 8 years seemed to conspicuously agree with the simple notion that under a global warming scenario we would expect less heat loss in the winter. This will reduce the likelihood of deep convection. In addition more melting of land and sea-ice will increase the fresh water content of the surface layers making it more difficult for deep convection to occur, even if the heat flux was unchanged.”

“While possibly true on multi-decadal or longer time scales, inter-annual to decadal climate variability will remain with us even as the planet warms. Very cold winters (albeit less frequent), will keep surprising us as they have in the past.” And they already do. The current winter is one of the coldest in Canada.

Igor Yashayaev confirms: “the 2008-2009 winter is going to be something! The upper mixed layer is already 0.5C colder and fresher than in any winter since the inception of Argo”. Yashayaev however challenges the authors of the article:

“I personally disagree that the last year convection caught us by surprise – as long as the ocean was getting salt and heat it was becoming ready to start loosing its excessive heat even in a moderate winter. It is like a boiling tea pot ready to release heat even if the air around is warm… Sooner or later the ocean would start mixing again, so combining this with 10-year record high heat loss, the 2008 kicked in!”

Was the 2008 event stronger than that of 2003? Yashayev says: “Yes! It was stronger and deeper in 2008, but it wasn't twice as deep; a stronger forcing led to a stronger convection, and, unfortunately, that is all we had.”

“The second surprise to the Vage et al. was the lack of any precursor event”, adds Martin Visbeck. “ There seemed no sign of a slow loss in ocean stratification, no increased upwelling in the middle of the gyre and no slow build up of stormier winters. I am not sure why we should expect that, given that all we know from middle latitude climate is, that it has almost a white spectrum. It is true that a preconditioned ocean will convect deeper compared to one with lots of fresh water on the surface. But if only the winter cooling is strong enough deep convection will kick in.”

“What I find very remarkable about the paper is our ability to do almost real time oceanography and close the real time attribution of interesting events. It is a great paper to read and a lesson for all of us who think that the trajectory to a warmer planet will be a smooth straight line without bumps along the road.”

Review for the EUR-OCEANS Consortium by Martin Visbeck, IFM-GEOMAR

REFERENCE: Våge, K., Robert S. Pickart, Virginie Thierry, Gilles Reverdin, Craig M. Lee, Brian Petrie, Tom A. Agnew, AmyWong and Mads H. Ribergaard 2009. Surprising return of deep convection to the subpolar North Atlantic Ocean in winter 2007–2008, Nature Geoscience, vol 2 JANUARY 2009

RELEVANT REFERENCES: Igor Yashayaev and John W. Loder, 2009. Enhanced production of Labrador Sea Water in 2008, GEOPHYSICAL RESEARCH LETTERS, VOL. 36, L01606, doi:10.1029/2008GL036162

RELEVANT LINKS: EUR-OCEANS Film ‘Gulf stream : The fall of a myth?’ ENG, FR

misc | about seo