Guest post: Ice cores reveal Antarctica is warming twice as fast as global average

As the world heats up in response to human-caused carbon emissions, scientists expect that the Earth’s poles will experience the largest increases in temperature.

This is already being seen in the Arctic, which has warmed nearly four times faster than the global average over the past four decades.

However, at the south pole, faster-than-average warming has only been reported in a few locations. This is partly due to the continent having relatively short and sparse temperature records, but also because it experiences large fluctuations in climate from one decade to the next.

In our new study, published in Nature Climate Change, we use ice-core records from across Antarctica to produce a reconstruction of temperature changes going back more than a millennium.

Our findings suggest that Antarctica is warming twice as fast as the global average, with around 20-50% more warming than estimates from the latest climate models.

Polar amplification

There are various reasons why the Earth’s poles are expected to warm faster than the global average, a phenomenon known as polar amplification. 

A key factor is the presence of so much snow and ice. These are extremely bright and so reflect a large proportion of the incoming radiation from the sun – far more than, say, the ocean or bare land. This property of reflectivity is known as “albedo”. 

When temperatures rise at the poles, causing snow and ice to melt, this reveals more of the land and ocean beneath. As a result, more sunlight is absorbed by the Earth’s surface, generating more warming. This is called a positive feedback loop.

Even if the warmer temperatures remain below freezing, as is often the case in Antarctica, changing temperatures and humidity can affect the characteristics of grains of snow. This snow “metamorphism” can change its albedo and, thus, how much sunlight it reflects.

However, the influence of albedo is just one part of polar amplification. A 2022 Carbon Brief guest post looked at the less talked-about factors.

The impact of polar amplification on the Arctic has been widely documented, but the picture is less clear for Antarctica. This has parallels with the changes in sea ice at each pole.

This is to be expected considering the two poles are very different. The north pole is an ocean surrounded by land, with relatively thin layers of ice floating on the ocean. In contrast, the south pole is a huge continent covered with several kilometres of ice and surrounded by an ocean. As such, we expect a much larger inertia of Antarctica’s climate than the Arctic’s.

As a result, while the signature of human-caused climate change over the past decades has been detected at the continental scale for Antarctica, it has not been consistently identified or quantified in observations.

Quantifying the warming in Antarctica

In Antarctica, the temperature increase has been hard to quantify until now because weather records for the continent are not as comprehensive as they are for much of the world.

For example, Antarctica has just 23 weather stations. In contrast, Europe – which is 25% smaller in area – has tens of thousands. And Antarctica’s weather stations have only been gathering data for the past 50 years, while some of Europe’s weather stations go back hundreds of years.

Of Antarctica’s 23 stations, just three are located inland. While two of them – the south pole and Vostok stations – have indicated an extremely large warming, at rates similar to those seen in the Arctic, these have not been matched with continental-scale temperature reconstructions. 

The lack of observations in the Southern Ocean and in Antarctica is a clear limit for scientists working on disciplines ranging from climate science to ecology, as illustrated by a recent call for action. 

To supplement the limited observations, scientists use reconstructions that are produced by combining weather stations data with climate model output. However, as these rely on weather stations that are predominantly in Antarctica’s coastal areas, this could be one of the reasons why continental warming has been hard to detect. 

The wind and oceanic currents that swirl around Antarctica could also explain why human-caused warming has not been clearly detected. For example, shifts to the north and south of the Southern Annular Mode (SAM) modulates the climate of the Southern Ocean and Antarctica and could hide the signature of climate change.

Ice core data

To overcome the lack of observed data, scientists can turn to proxy data – indirect records of climate imprinted on different parts of the biosphere and hydrological cycle. 

Ice cores are drilled out of ice sheets and glaciers to provide a cross-section of accumulated snow that has been compacted over thousands of years. The information contained in the ice includes dust from volcanic eruptions, air bubbles that provide samples of past atmospheres, and isotopes that offer evidence of past climates.

Ice cores from Antarctica provide a climate record going back almost a million years. 

In this study, we used a database of 78 ice cores gathered from across Antarctica by a large international collaboration led by the research programme Iso2k (PAGES). 

Combining these 78 ice cores, we reconstructed annual temperature changes over the past 1,000 years.

The interpretation of ice cores at such high resolution is challenging as snow in Antarctica is often blown over by the wind, which can add “noise” to the record, like static on an old videotape. In order to recover a meaningful climatic record for each individual annual ice layer, we needed to establish a strong link between the isotopes contained in each ice layer and Antarctica’s temperature.

Basic statistics were not sufficient here. We used dynamical system theory – a branch of chaos theory – to demonstrate that the current warming in Antarctica is between 0.22 and 0.32C per decade – outside the bounds of the continent’s natural variability over the last 1,000 years.

This is almost twice as strong as global warming, estimated by the Intergovernmental Panel on Climate Change (IPCC) between 0.14 and 0.18C per decade. This is also 20-50% larger than the estimates from the climate models used to produce the IPCC reports – even in East Antarctica, which was believed to be largely unaffected by climate change so far.

The impacts

The study also shows that in Antarctica, climate models underestimate not only the amplitude of global warming, but also that of natural variability. 

Improving the performances of models at the local or regional scale, without degrading their rather good performance at the global scale is a delicate exercise. Indeed, the fact that climate models underestimate human-caused warming in Antarctica suggest that the underlying mechanisms responsible for the polar amplification are not adequately represented in these models.

As predictions of future sea level rise, which rely on projections from the very same climate models, are sensitive to small changes in atmospheric temperature, this underestimation of Antarctic warming could be associated with underestimated projections of sea level rise – with dire consequences for low-lying lands. 

By quantifying the human contribution to Antarctic warming, this study provides yet another warning of the consequences of greenhouse gas emissions, even in the one of the most remote parts of the world.