CO₂ Fell So Antarctica Could Rise

A misconception that has been circulating around the internet since (I assume) long before I made my Twitter account, is that through Earth's geologic past CO₂ and climate change have never been connected. At least not in a way where CO₂ drives climate change. But any robust analysis of the geologic record reveals a pretty simple truth: CO₂ is an important "control" of climate change, and to deny that role is to leave major gaps in our understanding of the climate system. The gap I will be filling in this post is the initiation of permanent ice sheets in Antarctica some 34 million years ago. A time when CO₂ fell, and Antarctic ice expanded to the continental scale.

Let's avoid oversimplification. CO₂ is not the only driver of climate change. There are many, many others and they are very, very important. For example, many of the long-term (think on the scale of millions of years) climate changes seen in the geologic record cannot be explained without considering the tectonic changes to the world's continents and ocean basins. As continents move around like adrift puzzle pieces, they separate and join ocean basins. These changes reconfigure ocean circulation, and hence climate.

The birth of continental-scale glaciers in Antarctica was for a long time believed to be the result of the formation of the Antarctic Circumpolar Current ("ACC" for short). Today, this current is very important for keeping warm water away from Antarctica. 65 million years ago, Antarctica and Australia were connected like North America and South America are connected today. The water on one side of these combined continents cannot flow to the other side, inhibiting circumpolar ocean circulation. But after 65 million years ago the continents began to separate, eventually allowing for the circumpolar flow of water around Antarctica. Kennett (1977) concluded that this separation of Australia and Antarctica, forming the ACC, was the kick the climate system needed to grow sea ice in the Southern Ocean 34 million years ago, and permanent ice sheets 15 million years ago.

Kennett was not far off the mark with this theory. Around 15 million years ago marked the formation of the East Antarctic ice sheet. However, we now know from sediment cores sampled from near Antarctica that permanent glaciation began 34 million years ago (Barrett, 1996). What does this have to do with CO₂?

In 2003, DeConto and Pollard approached this problem from a modeling perspective. They wanted to see what effect, if any, atmospheric CO₂ had on glaciation in Antarctica. A good question to ask because in 2003, the reconstructed atmospheric CO₂ record had a gap from 40 million years ago to 25 million years ago -- which happens to be the timeframe of the study question.

Reconstructed partial pressure of CO₂ from 55 million to 0 million years. Note the gap in data. DeConto and Pollard (2003)

Pretty inconvenient, but the researchers pressed on. They considered three key factors: (in no particular order) the first being orbital cycles -- which are responsible for the waxing and waning of ice sheets in the current geologic period, the second being a gradual decrease in atmospheric CO₂ which they suspected was happening during that gap in data, and ocean circulation.

A benefit to modeling studies is the ability to control certain parameters of the climate system in order to test a certain scenario important to your study question. This is what DeConto and Pollard do in their study. With a decrease in CO₂, they modeled two Antarctic glaciation scenarios through a 10-million-year experiment. The first scenario, in their model, they did not allow for the ACC to form. In the second scenario, they did allow for the ACC to form. For both scenarios, atmospheric CO₂ concentrations fell.

Model runs in the DeConto and Pollard (2003) study. With the Drake Passage open (ACC formation) glaciation started early. With the Drake Passage closed (no ACC formation), glaciation started later. Glaciation are shown by a decrease in the change in sea level. From DeConto and Pollard (2003).

The authors of this 2003 study found that, as atmospheric CO₂ declined, permanent ice sheets formed in Antarctica in both scenarios. But, to the deserved credit of Kennett, CO₂ had to decrease further in the closed Drake Passage (no ACC) scenario than in the opened Drake Passage (ACC formation) scenario. This means that, for the formation of permanent ice sheets in Antarctica, CO₂ played a primary role; and the opening of the Drake Passage played a secondary (but still important!) role.

Of course, models are not everything. In 2003, key information about this time in Earth's history -- and Antarctic history -- remained a mystery. Modeling needed to be more robust, and we needed a better understanding of how CO₂ changed during this time.

In 2005, Pagani et al found that atmospheric CO₂ did decrease during this time, which was a very strong indicator that DeConto and Pollard were right. Then came Galeotti et al. (2016), which confirmed DeConto and Pollard's findings. Galeotti and colleagues studied an ocean sediment core off the coast of the East Antarctic ice sheet in the Ross Sea. Cycles in the type of sediment through the core reflect glaciation changes through orbital cycles. By seeing how these sediment cycles changed through the core, they figured out how extensive the Antarctic ice sheet was from 34 to 31 million years ago.

This 2016 study found that prior to 34.8 million years ago (high CO₂), Antarctica had a pretty small ice sheet. It did not extend all the way to the coast like it does today.  This smaller ice sheet was more sensitive to orbital cycles, waxing and waning according to amount of solar insolation it received. After 34.8 million years ago (low CO₂), however, Antarctica had a more extensive ice sheet and became less sensitive to orbital cycles. The authors of this study determined that at 34.8 million years ago, atmospheric CO₂ fell below 600 ppm, which they believe to be a threshold for extensive Antarctic glaciation.

Many things are responsible for climate change and the size of ice sheets. But to deny the important role of atmospheric CO₂ is to limit our understanding of the Earth's climate system. CO₂ is an important driver of climate change. And as we increase atmospheric CO₂ into the future, we threaten the ice sheets we have today.