How does deep water formation in the Irminger Sea relate to the North Atlantic cold blob?
In this blog post I’m giving some background information about our article that was recently accepted in Geophysical Research Letters. The article by myself and my NIOZ colleague Laura de Steur is titled “Strong winter cooling over the Irminger Sea in winter 2014-2015, exceptional deep convection, and the emergence of anomalously low SST” and can be found here.
What is deep water formation?
Deep water formation is an important process in the ocean circulation. It is the process by which warm, salty surface waters are cooled to the point where they become dense enough to sink and mix with the underlying deep water. The deep water formation forms the vertical link between the warm currents flowing pole ward along the surface and the dense, cold waters flowing equator ward in the deep ocean. This entire circulation is often portrayed as a conveyor belt (Figure 1), with deep water formation as one of the main mechanisms that drives the belt. Although it’s a very simplified picture, that sometimes get on our oceanographer’s nerves, we will stick with it for now.
Why do we care about deep water formation in the Irminger Sea?
Very simply put “the more wheels are driving our belt, the more stable it runs”. Much of our climate depends on the ocean circulation. The climate in northwestern Europe is relative mild compared to similar latitudes in Canada. The steady northward flow of warm water in the Atlantic is partly responsible for this. Although deep water formation is not the only process driving the belt, climate models show that if deep water formation were to stop it has the potential to alter or slow down the larger ocean’s circulation.
There are only a few locations where deep water formation occurs. On the southern hemisphere, deep water forms near Antarctica. On the northern hemisphere, deep water forms in the Mediterranean and around Greenland. These locations are unique because they fit a particular range of conditions needed for deep water to form. A local circulation that is just right and atmospheric conditions that can cool the water down enough to make it sink and mix (the Mediterranean is so salty that it needs to be less cold for this to happen). If the climate changes these conditions could possibly no longer be met. So the more baskets to put our deep water eggs in the better, so to speak.
Southwest of Greenland, in the Labrador Sea, deep water formation has been known to occur for a long time as there was a oceanographic weather ship moored right at the location of deep water formation in the 1940s. In the nearby Irminger Sea, southeast of Greenland, the weather ship was located in a less favorable position (closer to Iceland) and never observed deep water formation as extensive as in the Labrador Sea. The deep water seen in the southern part of the basin was thought to be brought there from the Labrador Sea by the ocean’s currents. It was not until the 1990s, when an early spring ship survey caught sight of newly formed deep water in the Irminger Sea, that the Irminger Sea was seen as a potentially important region of deep water formation. This was all described nicely in the paper by Pickart et al. (2003).
To test this theory, instruments were moored in the Irminger Sea by two groups. The Woods Hole Oceanographic Institution maintained two moorings from 2002 to 2004. Independently, the Royal Netherlands Institute for Sea Research deployed several moorings in 2003, one of which remains operated today. Initially the mooring’s time series shows some weak winters with mixing in the upper 400 m, which barely increased the density of the water. It was not until later in the record that proper deep mixing was observed. In the winter of 2014-2015 the deepest mixing recorded in the Irminger Sea was observed. Newly formed deep water was seen down to 1400 m and reached the same densities as deep water in the Labrador Sea.
Why is the timing of this record deep water formation important?
One thing that could disrupt deep water formation around Greenland is the input of meltwater into the ocean. Meltwater is freshwater, which has a much lower density than seawater. A layer of meltwater could form a barrier between the atmosphere that does the cooling and the seawater below that needs to be cooled to form deep water. In model experiment have been done where the melting of Greenland in a warming climate is simulated. The (strongly) enhanced freshwater input weakens the deep water formation and as a result the conveyor belt slows down. In these model runs the ocean around Greenland cools down as less warm water is transported north, while the rest of the globe warms under continued greenhouse forcing.
A similar pattern was seen recently in global temperature observations. Temperature anomaly maps of 2015 showed a cold northern North Atlantic in a warm world (Figure 3). This pattern led to speculation that the North Atlantic circulation had already slowed down. These were linked to other studies describing the increased melting of Greenland, which is indeed going on. However, from the measurements in the Irminger Sea (and similar measurements from colleagues in the Labrador Sea) it is clear that the deep water formation is still going strong. In fact, strong deep water formation events aren’t continuous, they are seen approximately every 10 years. This event fit the schedule as the last two events were in the early 1990s and 2000s. The extremely cold winter that forced the deep water formation is also responsible for the cold temperatures still seen there after winter. So in this case the pattern seen in Figure 3 is not due to a slowing down of the Atlantic circulation, but due to local interaction between the ocean and atmosphere. It is also important to note that we saw very little evidence of freshwater at the surface that might have originated from Greenland, so it’s quite likely that this process it not speeding up as much as feared (yet)…
This summer (27 July to 17 August) we are going back to the Irminger Sea on the research vessel Discovery. We will recover another year of data from the Irminger Sea moorings and do a hydrographic survey. It will be interesting to see whether the Irminger Sea has experienced another cold winter or if it’s slowly recovering to warmer temperatures again.
, & (2016), Strong winter cooling over the Irminger Sea in winter 2014-2015, exceptional deep convection, and the emergence of anomalously low SST. Geophysical Research Letters, 42, doi:10.1002/2016GL069596.
Pickart, R.S., Straneo, F., & Moore, G.W.K. (2003). Is Labrador Sea Water formed in the Irminger Basin? Deep Sea Research I, 50 (1), 23–52. doi:10.1016/S0967-0637(02)00134-6.