Jun 2019

Thwaites: Antarctica's Doomsday Glacier

Thwaites is the most important glacier in the world, which is why the National Environmental Research Council (NERC) and the USA’s National Science Foundation provided the funding for this research collaboration. It is the first time that scientists have been able to study the glacier up close, with the aim of using that data to make future projections of global sea-level rise.

Many countries invest heavily in sea defences. These offer some protection from rising sea levels, but a greater threat emerges from the increase in storm surges that occur as sea levels rise. Storms that breach sea defences and cause serious damage to the coastal infrastructure usually occur just once in a century. But in the next 100 years, there will be increases in both the vulnerability to, and occurrence of, these storms, due to rising sea levels. This diagram shows all of the contributing factors to sea-level rise on our planet:

An important factor is the thermal expansion of the oceans as they absorb the extra heat coming in from the atmosphere from greenhouse gases. That thermal expansion will continue, even with major efforts to reduce emissions because the existing greenhouse gases in the atmosphere will continue to warm the planet for an extended period. However, the rate of warming in the ocean (and land) can be reduced by these measures. There is also the melting of ice on land - glaciers around the world are losing ice into the ocean which is contributing to sea-level rise (see Questions section below). However, it is the ice sheets in Antarctica and Greenland that will have the biggest contribution in the next 100 years and more distant future.

The Inter-Governmental Panel on Climate Change (IPCC) has been examining this problem, in a series of major reports spanning many years. A follow-up to their 2013 assessment report on the oceans and cryosphere is due to be published soon, but little change is expected from the projections below:

These 2013 projections suggest that if we do everything we can to reduce greenhouse gas emissions we can expect only 30cm of sea level rise (blue), which is a continuation of what we have seen over the past 100 years. This contrasts with the projections for a high emissions future (red) which show sea-level rise of a metre or more. The high and low emissions scenarios overlap, firstly because sea-level rise is already happening – no matter what we do about new carbon emissions it cannot be halted. Glaciers will continue to melt and the oceans will continue to absorb heat and expand. Secondly, scientists don’t fully understand how to forecast the impacts on our ice sheets and glaciers, especially in Antarctica and Greenland, as they respond to both warming air and increased ocean heat.

What does this mean for us? Estimates suggest that 75% of the world’s coastlines will experience sea-level rise by the year 2100. In the UK, Scotland’s ice sheets disappeared naturally around 11,000 years ago. Adjustments by the Earth’s crust to the loss of weight on the landscape has resulted in the south of England sinking into the ocean by a couple of millimetres each year. In London, using the lower projection of 30cm sea level rise and adding the 20cm of subsidence that will occur in that time period, we could expect to see half a metre of sea level rise at the Thames Barrier by 2100. The Barrier was designed to provide a level of protection that would only be breached once in every 200 years. With 50cm of sea level rise that protection will be reduced, meaning that a storm would breach the current defences at least once in an individual’s life span in the city. If sea levels rise by a metre (which is still within the 2100 projections on the graph above) the existing defences will be breached on average every 10 years. The Environment Agency relies upon accurate projections of sea-level rise to plan and invest in new sea defences that will protect the capital.

The biggest uncertainties lie in the ice sheets - Antarctica and Greenland. Because there are only two of them, statistical projections are extremely difficult so scientists need to understand what’s driving change in those systems. In Antarctica the ice is being lost from the areas in red on the map below:

Most of Antarctica maintains a balance – the amount of icebergs produced every year is replaced by the same amount of snowfall – except the areas in red. Near the centre of the largest red area is the Thwaites Glacier region. Thwaites is the largest glacier in the area and this is where the International Thwaites Glacier Collaboration (ITGC) will be focused for the next few years. The team are trying to find out in detail how much ice Thwaites is losing, and how this might change in the future. In particular they’re seeking to understand how the interaction between the glacier and the sea is changing – any small changes around the edge could be magnified with a progressive inland glacial retreat. The second factor that they’re concerned about is how Antarctica produces icebergs in the future, perhaps changing from producing big, tabular icebergs every few years, to Greenland-style icebergs that are produced every day during the summer season.

Professor Karen Heywood - How robots and seals investigate glacier retreat The ocean is a major factor in ice mass loss and this is most evident in West Antarctica where the ocean is warmest and comes closest to the continent. TARSAN (The Thwaites-Amundsen Regional Survey and Network) is one of eight different projects that are being run concurrently as part of the ITGC, and its focus is studying the ocean, ice shelf and atmosphere.

The TARSAN team (a collaboration between the UK, USA, Sweden and Korea) has been studying the temperature and salinity of the water that goes under the floating ice shelves at the coastline of the Thwaites region. But it’s an extremely difficult place to observe, especially during the Antarctic winter, when sea ice covers the entire region.

With the TARSAN project, the oceanographers teamed up with biologists to use seals to collect this elusive data. Tags were glued onto the heads of seals so that, when the seals dive to the seabed to feed, they can take measurements of the temperature of the ocean at the same time. Seals are able to find gaps in the sea ice during winter so measurements can be recorded all year round. The tags do not change the seals’ behaviour and are glued to their fur, meaning that when they moult the tags fall off naturally. In fact, the tags also assist biologists in understanding how the seals survive in this remote area.

A 2014 project in the Eastern Amundsen Sea collected 10,000 full-depth profiles of temperature and salinity from tags on seven elephant seals and seven Weddell seals from February to October. Data of this quantity had never been collected in the Antarctic winter before and was considered revolutionary.

The ITGC team decided to use the same techniques to collect data in the Thwaites region and will be tagging seals to measure ocean temperatures over the next three years. Twelve seals are already hard at work, transmitting data back to the UK in real time, right now.

It’s too dangerous to get a ship close enough to take measurements from the calving ice front (where the icebergs break off), so the TARSAN team are using autonomous underwater vehicles (AUVs) to explore the area beneath the ice shelf. The vehicles travel below the floating ice and take measurements of the relatively “warm” water at the boundary between the ocean and the ice. Following a successful trial this winter, they aim to send two propeller-driven AUVs – one of which is the famous Boaty McBoatface - underneath the Thwaites Glacier in 2021. The robots have various sensors onboard and will be able to measure water properties, current velocity, heat flux, and turbulent mixing processes.

The team are also using buoyancy-driven ocean gliders to understand the processes taking place at the glacier front. The gliders are relatively cheap, efficient and last a long time so they are able to create top-to-bottom transect profiles of the ocean in the area adjacent to the Thwaites Glacier. Much of this technology, and some of the techniques, have never been used in this area before so the scientists are hoping to get a vast data set that they can use to make the most accurate predictions yet about melt in the Thwaites region and its impact on the rest of the world.

Professor Doug Benn - Computer Models and Sea-Level Rise from West Antarctica The DOMINOS and PROPHET projects within the ITGC aim to use computer models to predict how the Thwaites Glacier will evolve over time. These models simulate the behaviour of the ice sheet and how it changes, allowing scientists to conduct experiments with different future scenarios.

Warm water under the fringes of the ice sheet is causing melting and, as the ice gets thinner, it becomes more likely to float and will move faster. The ice sheet bed in West Antarctica is very deep - well over 1km below sea level in many areas - which means that if the ice gets thin enough it will float, break up and collapse.

Models can be used to see the effects of higher ocean temperatures. If the temperature is increased by just one degree, the Thwaites Glacier shows a rapid increase in melting, higher rates of ice flow and greater ice discharge into the ocean. Floating ice is also lost from large calving events, creating breakaway icebergs. The effects of the combination of ice sheet melting and calving, as shown in De Conto & Pollard’s model below, are described as “frightening” by Professor Benn.

In De Conto & Pollard’s model, the top map shows what will happen if the Paris Agreement targets are met, the middle map shows what happens if the CO2 targets are exceeded slightly, and the bottom map shows what will happen if CO2 continues to rise at present day rates. Even a moderate increase in ocean and atmospheric temperatures will have a dramatic impact on West Antarctica, changing the outline of the continent entirely. The DOMINOS and PROPHET projects are refining these models, performing a series of experiments to understand more about how Antarctica will respond to different greenhouse gas scenarios. Under high carbon futures (i.e. exceeding the Paris Agreement targets), the models predict a very rapid loss of ice due to increased melting beneath the ice sheet, faster ice flow, calving and ice instability.

But the key question facing scientists and decision makers is exactly how much ice will be lost from Antarctica, and how quickly? This will have an impact on global sea levels, so how much can we prevent it, and how much do we need to invest in mitigation? Professor Benn warns that the more we fail to prevent, the more we will have to mitigate.

Professor Ted Scambos – How the ITGC Will Help Us Understand the Future

Thwaites Glacier is the size of Great Britain - or the US state of Idaho. Imagine a layer of ice about 1.5 – 2km thick resting on top of that, then picture that vast mass, over the next few centuries, moving into the ocean.

Scientists in the US and UK came to the realisation at the same time that Thwaites was the most important place to study. This provided a lot of momentum for funding and energised the science community. The ITGC programme is highly complex and consists of eight projects that are designed to be collaborative, to collect and share data at the same time, and feed that data into modelling programmes.

The logistics involved in getting 80 people out to such a remote part of Antarctica are nearly as complex as the science. Thwaites is roughly equidistant from the US station at McMurdo and the British research station at Rothera. Tractor traverses from both stations are used to get all of the material and the people to where they need to be, and no journey is wasted in terms of scientific activity. The vehicles take measurements of the ice sheet en route, using sensors, seismology and radio echo technology to measure ice thickness and ice layering. The UK and US have invested heavily in Thwaites because melting in Antarctica is likely to have a 15 – 30% greater impact on sea level rise in the northern hemisphere than in the south. This is because ice that flows into the ocean does not get distributed equally as sea levels rise around the world’s oceans - it’s not like raising the level in a bath. With so much mass moving off the continent and into the ocean, the shape of sea level on Earth changes and sea levels in the area close to the glacier will actually drop.

The Antarctic ice sheets are trillions of tons of mass and, with a large amount of mass, comes a lot of gravitational pull. The ice sheet mass pulls the ocean up against the side of the continent, but if that ice has melted and the mass has reduced, the gravitational pull is weaker and the ocean relaxes away from the coastline. This is why the northern hemisphere, far away from the source of the ice loss, takes the brunt of the sea-level rise.

Melting in Greenland has the opposite effect - and in fact the UK is slightly protected from sea-level rise if the loss is mostly in Greenland. However, with global warming, both Antarctica and Greenland are likely to lose a lot of ice, impacting the UK and US, but mid latitude areas and the tropics will suffer most with sea-level rise.


How fast is the glacier migrating inland at the moment?

Thwaites is retreating at a rate of hundreds of metres per year, and in some places it’s more rapid than that. It’s worrying because as the bed gets deeper further inland, this process is likely to reach a tipping point.

With a sea-level rise of 1.2 – 1.4m around the US, what impact will this have on natural catastrophe rates and extreme weather patterns?

The impact of weather patterns due to ice sheet loss will be quite localised to that part of Antarctica because the Southern Ocean is a fairly isolated system in terms of how the winds and ocean currents flow. Increases in storm action will be triggered more by changes in sea ice extent and Greenland warming and melting. With Arctic sea ice loss, there is more exchange of heat and moisture between the air and the ocean in the far north. This will impact on the very complex meteorological system of the Northern Hemisphere. Similarly, loss of ice in Greenland will affect ocean circulation in the North Atlantic, indirectly affecting weather and climate in Europe. However, the sea-level rise from loss of the ice sheets will make the impact of those storms more severe.

Do the melting glaciers in the Himalayas and Iceland have an impact?

In combination, all of the mountain glaciers around the world could produce about 0.5m of sea level rise, however, not all of that ice melt will reach the ocean, nor will it melt in the current century. Models of glacier melt suggest a maximum of 20 cm sea level contribution by 2100.

As a result of this programme, will we have a better understanding of the implications of black carbon and surface algae that we’re currently concerned about in the Arctic?

Glaciologists tend to be northern or southern specialists, but Greenland is a great analogue for how parts of the Antarctic might be after 100 years of warming. The difference between Greenland and Antarctica is that half of the ice loss in Greenland is from the surface, whereas that figure is almost zero for Antarctica because the climate is much colder. In a future, much warmer world, surface melting will affect the coast of Antarctica more widely and lessons learned in Greenland will be extremely valuable.

With the ITGC, did you start with a thesis that you then set out to prove? Was the process created around a prediction or set up as an experiment?

The idea that Antarctica will become unstable as a result of global warming was first presented as a theory in the 1970s by the glaciologist John Mercer. He said that the glaciers on the Antarctic Peninsula would be the canary in the coal mine, indicating instability for West Antarctica. And so it has come to pass. This project began with a theoretical idea, but satellite technology allowed us to see these things happening and led us to the understanding we have now.