Another obscure rapid climate change event: the Younger Dryas Cooling

We have talked about two distinct repetitive abrupt climate changes that are evident in the ice core records over the last 60,000 years: Dansgaard-Oescher (D/O) cycles and Heinrich events. 

However there is one particular cold event in the recent geological past that scientists are not able to identify as either one of them. This is called the Younger Dryas cold event (YD).

(Rahmstorf, 2002)

The YD cold event appears as an interesting rapid climate change event, since it has been detected during a phase of long-term warming. So why did climate all of a sudden drop to nearly ice-age like conditions? The interesting part is that geological evidence strongly suggests first a cooling, then a warming of roughly 4-8 degrees in Greenland within less than 20 years (Barber et al., 1999; Alley et al., 1993; Mayewski et al., 1993; Dansgaardet al., 1989)! 

The YD does indeed deserve the name RAPID climate change…

The most obvious mechanism known so far to drastically change northern hemispheric climate on such a short timescale is again the Atlantic circulation. In 1989, Richard Fairbanks researched coral cores from Barbados and calculated melt water discharge into the Atlantic Ocean from oxygen isotope records in the corals (see Science Fact for more information). He found two distinct events during the YD where freshwater entered at a maximum rate of 14 000 and 9 500 km3/year! To put that in relation: during Heinrich events models had calculated 1.25 million km3 of freshwater release, but in more than 250 years (see last post). That is a maximum rate of only 5000 km3/year. Today, the two big rivers flowing into the Atlantic (Mississippi and St Lawrence rivers) release barely 900 km3/year (Fairbanks,1989).

Hence, it is extremely likely that such a flush event of freshwater would have an effect on the thermohaline circulation in the Atlantic.

The question is what could cause such a sudden release of freshwater in such a short time?

In the late 1980s, Broecker and colleagues concluded that most likely a large freshwater lake emptied into the North Atlantic Ocean (Broecker et al., 1989; Broecker et al., 1988). They knew that the big Laurentide ice sheet on North America had been melting for a while (remember we are actually in a time of climate warming). Due to big moraines in the landscape it was possible to form two large lakes, Lake Agassiz and Lake Ojibway, located near the Great Lake area of today (Barber et al., 1999). Broecker and colleagues noticed that the normal outflow of the lake went southwards through the Mississippi river basin into the equatorial Atlantic. However at the same time of the YD onset, the outflow of the lakes changed, most likely due to a moraine breakdown that had served as a dam. Large amounts of water now catastrophically raced through the Hudson Straight into the North Atlantic (Broecker, 1989). A new study soon to be published in 2015 by Li &Piper shows that the Labrador Current (flowing along the North American coastline) did indeed speed up during the YD event probably due to all the extra freshwater input.

These findings led to a great increase in research about the YD and possible routes which the water could have taken. In 2010, Murton et al. presented evidence for old fluvial layers deposited in the Arctic Ocean. They argue that there was a second important outlet which formed during the same time and enabled large freshwater floods straight into the deep water formation sites. Hence, the Hudson Straight outlet might have been of only secondary importance in providing catastrophic freshwater flushes.

Outflow via Mississippi River and via Hudson Straight (Eastern Outlets) during the Younger Dryas cold event. notice that y-axises are inverted! (Broecker et al., 1988)

Both outflows from Lake Agassiz and Lake Ojibway (eastern & northern red arrows) 

(Murton et al., 2010)

In general, scientists are sure that this event led to a shutdown of the Atlantic circulation (Clark et al., 2001; Rahmstorf, 2002; Barber et al., 1999). Especially the release of freshwater directly into the areas of deep water formation, as it is the case with the Arctic outflow (Murton et al., 2010) will lead to fast reorganisations of  the circulation and stop new deep water formation if the freshwater force is strong enough.

Yet, there are many unresolved questions that scientists are still trying to work out.

One question deals with the mechanisms that can make a rather local disaster a global climate event. Here Rach et al.(2014) just found evidence that there was an obvious delay between climate cooling in Greenland and in Western Europe. This delay may show that not only cooling temperatures, but also changes in wind and rain patterns have influences on terrestrial climate (Rach et al., 2014).

Another question looks at CO₂ evolution during the YD. With the YD being a cold event, one would have expected low CO₂ concentrations. However, Steinthorsdottir et al.(2014) show an abrupt increase, then decrease of CO₂ at the beginning of the YD period suggesting that something had changed in the ocean circulation and forced it to “burp out” a cloud of CO₂. Did that change in ocean circulation maybe influence the YD event? We don’t know yet.

Short review:

After looking at three different distinct abrupt climate change events (Daansgard-Oescher cycles, Heinrich events, the Younger Dryas cooling event), we see that ocean circulation is not necessary the driver, but the amplifier of rapid climate changes. It almost seems like the Atlantic Ocean is operating in several modes... and what about today....?

Keep reading the blog ;)

Dansgaard-Oescher cycles and the Thermohaline Circulation

Let's have a look at the last record again:

(Rahmstorf, 2002)

We see that all Dansgard-Oescher (D/O) cycles happen during the big glacial period. So one hypothesis that scientists came up with believes that the triggering of the cold drops could be due to ice bergs. The large amounts of ice rafted debris (IRD, see last posts info box) on the ocean sea floor suggest that big pieces of sea ice broke off the big Laurentide ice sheet sitting on top of North America and the big Scandinavian ice sheet located on top of Scandinavia and Northern Europe, and floated southwards on the Atlantic Ocean. Of course ice sheets are much colder than liquid water, so melting must have happened fast. Hence, large amounts of freshwater were released to the North Atlantic changing the salt content within the water.

From the last posts, we know this implies a reduction of North Atlantic deep water formation and hence a weakening of the whole Atlantic THC!

Models indeed show that input of freshwater to the North Atlantic leads to a weakening of the Atlantic circulation. However, many factors, such as amount, rate and location of freshwater input, seem to influence how drastic this weakening is (Ganopolski &, Rahmstorf, 2001; Clark et al., 2002; Hu et al., 2008). Also, many model runs suggested other earth compartments to play a role in the global distribution of cooling by changing wind, rain and evaporation patterns (Clark et al., 2001).

So, sadly the answer is not so simple. If models show something, does that mean it really happened?

Blunier et al. (1998) may have found the missing link when they were comparing ice core records in the Arctic with the Antarctic. They simply plotted both temperature curves on the same time scale (not as simple if you have to do it….) and saw that both records unexpectedly did not line up. When the arctic temperatures were cold, the Antarctic temperatures were warm and vice versa. This was not the case for all D/O cycles, but very distinct for few. How could that be possible?

During the same year, Stocker (1998) proposed the solution: the Atlantic circulation. We all profit from the heat the Gulf Stream constantly transports to the north. However, we tend to forget that the heat is actually stolen from the southern Hemisphere. If we compare heat transports in other oceans, heat north of the equator goes north, while heat south of the equator goes south. Contrary, in the Atlantic heat goes north no matter where it is located, due to the THC.

What Stocker is implying for the D/O events is that a cooling in the North Atlantic will lead to more ice bergs melting and more freshwater input. This will weaken the THC and slow down heat piracy from south to north. As a result, the southern hemisphere will end up with more heat, leading to a warming in the south, while there is a cooling in the north. This process of the thermal bipolar-see saw (or sea saw) can be found in actual climate models (i.e. Stenni et al., 2011) as well as in climate models (Seidov& Maslin, 2001).

As a summary we can conclude that obviously Atlantic THC played a major role in forming the D/O cycles. 
However what actually induced climate to change is still discussed. Some say solar insolation gave the first initial forcing (i.e. Cruz et al., 2005), some say the changes in Atlantic circulation can explain the climate changes (Seidov& Maslin, 2001).  
Still, then what changes the THC? 
Ice volume… what changes ice volume? ..... You see the problem.

...

Why are we interested in ocean circulation? A short trip to the Earth's past.

Many people today are worried about the future of the earth when thinking of climate change. To most humans, climate and natural settings were considered the one stable concept in their otherwise unstable life. Hence, the thought of a 4˚C warmer Earth, as it is postulated by the IPCC and many newspapers often triggers communal fears of mass extinction that may even include the human race.

You may panic.

http://www.demonsinthedark.com/panicandalcohol.htm

However! Here is the interesting bit. During the last century, climate scientists have found evidence for long term climate changes throughout the earth’s history (e.g. Daansgard et al., 1982; Daansgard et al., 1993). Proxy data (eg by converting information found through biological and abiological indicators into climate variables) or even direct measurements (eg trapped air bubbles in ice cores) in geological records have shown that Earth has went through climates much more aggressive than the one we know.

-          Precambrian (600-3500 Ma): Snowball Earth: The whole of the earth’s surface was covered in ice (Hydeet al., 2000; Donnadieu et al., 2004)

-          Cretaceous (65-144 Ma): Greenhouse Earth: Ice did not exist. Fun fact: CO₂ pressure was above 1000 ppm vs today at 400 ppm (Barron et al., 1981; Barron,1983; Kuypers et al., 1999)

This shows that the Earth has a much greater climate span than we would have initially believed. Since our climate today includes ice on Earth (glaciers and permafrost) we are actually closer to the snowball earth climate than the Cretaceous version; meaning, we are stuck in an Icehouse Climate, while an ice free earth represents a Greenhouse Climate.

To get a more accurate understanding of our current climate situation, we should take a closer look at the recent past of our climate.

Vostok Ice core time series and Insolation, taken from Petit et al. (1999)

Petit et al. (1999) have managed to characterize climate for the last 400,000 years. We see that temperature (line b) has fluctuated strongly between cold glacial and warm interglacial periods. The Insolation curve (line e) suggests that those fluctuations are initiated by changes in solar energy on earth. Thanks to Milankovitch, we know today that climate over the last ice ages was indeed forced by three external variables (Eccentricity, Obliquity and Precession) with different oscillations (Zachos et al., 2001).

Now put yourself in the shoes of a climate change scientist and try to answer the following question using the graph above:

What is the difference between warming phases and cooling phases?

Post your answer if you like and see whether you were right next time J