Exploring environmental changes in the Antarctic with Blue Planet II

I joined the BBC Blue Planet II expedition to the Antarctic, where we made the first dives in minisubmarines to reach 1 km deep there. As a deep-sea biologist, those dives gave me new insights into how “dropstones” (typically car-sized boulders that fall from passing icebergs) shape the pattern of life on the ocean floor around Antarctica.

But the expedition also provided an opportunity to collect data about the environment – in particular, the temperature and salinity of water at different depths. Whenever we explore the oceans, I think it’s important to collect all the data we can, especially when we’re visiting remote places such as the poles and the deep ocean.

 

 

The environmental changes taking place in the Antarctic are complex – an intricate interplay between ocean, atmosphere, land ice and sea ice – with different Antarctic regions changing in different ways (e.g. Steig et al., 2009). It’s not a simple story of “more carbon dioxide traps more heat in the atmosphere that then melts the ice” (the changes in the Arctic are perhaps more straightforward in that regard, as this summary by David Shukman, BBC News Science Editor, explains).

That’s why thousands of scientists have been investigating different aspects of what’s happening in Antarctic over decades, to untangle the complex knot of processes involved.  And as well as interacting with each other, the strands of that knot (ocean, atmosphere, land ice, sea ice) are also affected by natural variability (e.g. Mulvaney et al., 2012), can sometimes be influenced by changes elsewhere on our planet such as the equatorial Pacific (e.g. Ding et al., 2011), and by changes that we have made to the atmosphere – not just adding carbon dioxide to it overall (e.g. Marshall et al., 2004; Goosse et al., 2008), but also through the depletion of the ozone layer above Antarctica affecting winds around the continent (e.g. Thompson & Solomon, 2002).

The complexity of what’s happening in the Antarctic is all the more reason to take every opportunity to collect more data points, like these during our expedition:

 

 

On their own, our measurements of ocean temperature and salinity at different depths during the Blue Planet II expedition don’t tell us much – of course you can’t really understand processes or measure changes from one set of measurements in one place at one time. But the expedition’s data are archived so that scientists can compare them with other measurements, taken at other times and in other locations and combined into large datasets by international research programmes, to help towards building a more detailed picture of what’s going on.

For Blue Planet II, we were working at the northern tip of the “Antarctic Peninsula”, the narrow finger of land that points up towards South America. What’s clear from previous studies is that this particular part of Antarctica has seen some rapid recent changes, such as:

• an overall increase of around 1.8 degrees C in annual average air temperatures since the late 1940s, recorded at Argentina’s Esperanza Base (data available here), which was the closest weather station to where we were working:

• …and that’s consistent with the general pattern of warmer annual average air temperatures recorded by other weather stations towards the tip of the Antarctic Peninsula over that period (Vaughan et al., 2003).  In the last few years some parts of the Peninsula have seen cooler summers again (Turner et al., 2016), possibly as a result of the regeneration of the ozone layer above Antarctica affecting the winds around it, but that doesn’t reverse the large surface air temperature increase in this region of Antarctica since the mid-twentieth century;
• Esperanza Base also holds the record for the warmest surface air temperature ever recorded in mainland Antarctica and its surrounding islands: 17.5 degrees C on 24 March 2015. That one-off warm day was probably caused by a “Foehn Wind” (which we’ll explore later);
• an increase of more than 1 degree C in summer sea surface temperatures around the tip of the Antarctic Peninsula since the 1960s (Meredith & King, 2005);
• the retreat of 88% of the glaciers on the Antarctic Peninsula since the 1940s (Cook et al., 2005; Cook et al., 2016), and the total collapse of floating ice shelves such as Larsen A in 1995 and Larsen B in 2002 (Cook & Vaughan, 2010).

Some of the changes on the Antarctic Peninsula may lie within the range of natural variations recorded in ice cores over the past 2000 years, but the recent warming is unusually rapid (Mulvaney et al., 2012). And the last time that any changes like these occurred, floating ice shelves such as Larsen B did not collapse (Domack et al., 2005), and there weren’t 7.6 billion of us who could be affected by what is going on. That’s what takes us into uncharted territory today.

 

Glaciers, floating ice shelves, and sea level rise

Sea level rise is one change that can affect us all, and it involves changes in the rate at which ice forms on land and flows into the ocean. Snow that falls on Antarctica eventually compresses under its own weight to form ice, building up ice sheets on the land. As more ice builds up, it flows out in glaciers and ice streams towards the coast, where it then floats out over the sea, forming the floating “tongues” of glaciers and more massive floating ice shelves. Eventually, the ice breaks off those floating tongues and shelves to form icebergs of fresh water, sometimes hundreds of metres thick.

The “calving” of icebergs from the ends of glaciers is a natural process, as the amount of snow falling on land gradually builds up more ice over time. During the Blue Planet II expedition, I was particularly interested in how that process delivers “dropstones” to the ocean floor – rocks scoured from the land by the glaciers and carried out to sea on the underside of icebergs, from which they fall to provide “islands” of rocky habitat for different types of animals on the seafloor.  Glaciologists can monitor numbers of icebergs and the flow of glaciers using satellites, but I also had an opportunity to make some closer observations of bergs from the air, to try to understand more about how they deliver dropstones that we saw on the ocean floor during dives.

 

 

But when it comes to sea level rise, what’s important is the rate at which ice flows from the land into the sea via this process. The floating ice shelves and floating ends of glaciers act as buttresses for the ice, slowing its slide into the ocean.  The buoyant force exerted by the ocean on the end of the floating “tongue” of a glacier, or across the front of a floating ice shelf, pushes back against the downhill flow of ice from the land.  If the floating glacier tongue or ice shelf becomes thinner, or retreats so that the “grounding line” where it starts to float out from the land is further inshore, then it offers less resistance to the flow of ice from land into the ocean (there’s a really good diagram explaining the forces involved here).

So if the floating “tongues” of glaciers or more massive floating ice shelves become thinner or disintegrate, then ice flows more rapidly from land into the sea. Because ice on land is not yet afloat, when it reaches the sea it pushes up the water level, like dropping ice cubes into a drink. In contrast, “sea ice” – the skin of ice only a few metres thick created when the surface of the ocean freezes – does not have the same effect on sea level, because it’s formed from water that is already in the ocean.  What matters then, for Antarctica’s role in sea level rise, is how the floating tongues of glaciers, floating ice shelves, and flow of ice from land to sea are changing.

 

Antarctica in three parts

Antarctica is huge: its land area (~14 million km²) is about 50 percent larger than the United States. To understand what’s going on for sea level rise, we need to divide Antarctic into three parts, because what’s happening is different in these regions:

• the Antarctic Peninsula, which is the narrow finger of land that points up towards South America (and we were working at the tip of the Peninsula for the Blue Planet II expedition);
• the area covered by West Antarctic Ice Sheet, and in particular the coastline on the Amundsen Sea where several glaciers drain that ice sheet into the ocean;
• the area covered by the much more massive East Antarctic Ice Sheet, which covers most of the interior of the continent and the South Pole.

 

 

Fortunately, there isn’t that much ice on land on the Antarctic Peninsula, where air and sea temperatures are changing the most, to drive much sea level rise from there. Further south, the West Antarctic Ice Sheet has the potential to drive much more, and some of its floating ice shelves are thinning and retreating, because of an upwelling of warm deep water. Meanwhile the enormous East Antarctic Ice Sheet fortunately appears stable, though one particular glacier outlet could be changing in ways similar to some of the glaciers draining the West Antarctic Ice Sheet.

 

Changes to glaciers draining into the sea on the Antarctic Peninsula

596 out of 674 (i.e. 88% of) sea-ending glaciers on the Antarctic Peninsula have retreated since records began in the 1940s (Cook et al., 2016), and 7 out of 12 of its floating ice shelves have retreated significantly or collapsed completely (Cook & Vaughan, 2010). The retreat of glaciers and floating ice shelves on the Antarctic Peninsula is driven by two different processes, one operating on each side of the Peninsula. On the western side, the floating ends of glaciers are mainly being eroded from underneath by an upwelling of warmer deep ocean water, rather than changes in air temperature. But on the eastern side of the Peninsula, more frequent warm winds create melt ponds that eat through floating ice shelves from above.

Most of the retreating glaciers of the Antarctic Peninsula are on its western side, along the coast of the Bellingshausen Sea, where there has been upwelling of warmer “Circumpolar Deep Water” to depths of 100-300 m (Cook et al., 2016). That upwelling of warmer deep water to melt those glacier tongues is a different process to the warming of sea surface temperatures further north on the Peninsula – and not a result of the warming air temperatures on the Peninsula either – but instead driven by changes in wind patterns around Antarctica that in turn drive some changes in ocean circulation (e.g. Peck et al., 2015).

 

 

 

Meanwhile, and in contrast, on the eastern side of the Antarctic Peninsula, the demise of the Larsen A floating ice shelf in 1995, and the Larsen B floating ice shelf in 2002, was probably hastened by warm Foehn winds, which cause melt ponds to form on the surface of floating ice shelves, eventually eating through them and breaking them up.

Foehn winds are warm winds created when a wind drops down the side of a mountain, compressing the air and thereby warming it. On the Antarctic Peninsula, circumpolar winds get pushed up against the western side of the mountains of the peninsula, and when they spill over the top or through gaps to the eastern side, they drop down and create the Foehn warming effect there, contributing to the break-up of floating ice shelves (Cape et al., 2015). There’s a useful BBC News feature about the phenomenon – and its effect on ice shelves on the eastern side of the Antarctic Peninsula – here.

(As an aside, the factors behind the huge iceberg that recently broke off the Larsen C ice shelf, and whether the remainder of that floating ice shelf will similarly collapse soon, are even more complex, as discussed here).

Immediately after Larsen B ice shelf collapsed in 2002, scientists measured an acceleration in the slide of ice from the land there into the sea, contributing to sea level rise (Rignot et al., 2004). Fortunately, the Antarctic Peninsula is only a narrow strip of land and there’s not so much ice on land there: if all the ice on the Antarctic Peninsula slid into the sea, it would raise global sea level by about 24 centimetres (Pritchard & Vaughan, 2007), and what really matters is how quickly that process happens.  To put that in context, the current rate of global sea level rise is about 3 millimetres per year, and ice loss from the Antarctic Peninsula is currently contributing around 0.16 mm per year to global sea level rise (Pritchard & Vaughan, 2007).

 

Changes to glaciers draining ice into the sea from the West Antarctic Ice Sheet

Further south, the Western Antarctic Ice Sheet contains much more ice locked up on land than on the Antarctic Peninsula, and changes in the much larger glaciers draining it into the Amundsen Sea (e.g. Pine Island Glacier and Thwaites Glacier) are the main concern here. If those glaciers fail as buttresses, the impact on sea level rise could be much greater than from the Antarctic Peninsula.

The West Antarctic Ice Sheet is also a “marine ice sheet”, which means that its base, although inland, actually lies below sea level.  This may make it particularly susceptible to a “runaway” effect if the ice shelves buttressing its flow into the ocean weaken, potentially allowing the ice that is currently inland to become afloat and thereby drive up sea levels substantially (Joughin & Alley, 2011).

The floating end of Pine Island Glacier already appears to be thinning and retreating (e.g. Wingham et al., 2009), as a result of warm deep water upwelling to melt its underside (Jenkins et al., 2010). As a result of such changes, the amount of ice sliding from land into sea from the Western Antarctic Ice Sheet has increased, and already contributes around 10% of the observed rise in global sea level (Rignot et al., 2008). If all the ice in the West Antarctic Ice Sheet ended up in the ocean (perhaps unlikely, though it has declined substantially in the more distant past, e.g. Pollard & DeConto, 2009), there’s enough ice there in total to raise global sea level by around 3.3 metres (Bamber et al., 2009).

 

 

Changes to the East Antarctic Ice Sheet

On the other side of the Antarctic continent, the even larger East Antarctic Ice Sheet contains yet more ice on land – more than ten times the amount locked up in the West Antarctic Ice Sheet. The good news here is that it appears to be stable – in fact, increased snowfall over central Antarctica means that the East Antarctic Ice Sheet is actually gaining mass at the moment (King et al., 2012). And where it drains into the sea via glaciers, there aren’t upwelling warmer waters flowing close inshore to erode the floating ends of the glaciers and the floating ice shelves that buttress the ice flow.

There is one exception, however: at the Totten Ice Tongue, there are seafloor channels bringing warm deep waters to the underside of the floating end of the glacier, causing it to get thinner (Rintoul et al., 2017). That may be unique for a glacier draining the East Antarctic Ice Sheet, but the “catchment area” of land ice for the Totten Ice Tongue is almost as large as the whole West Antarctic Ice Sheet, so changes in Totten alone could make quite a contribution to sea level rise (and there’s a useful Nature news feature about the “sleeping giant” of Totten here).

 

How does all this affect us?

Changes in the Antarctic are not the only factor contributing to global sea level rise, of course. Roughly half of the currently observed sea level rise (which is around 3 mm per year total at the moment) comes from thermal expansion of the ocean – as the ocean absorbs heat from the atmosphere, its volume of water expands as it gets warmer.  And changes in other ice sheets and glaciers elsewhere in the world, such as Greenland, also contribute to sea level rise.

Predictions of sea level rise by the year 2100 for the Intergovernmental Panel on Climate Change are almost as much as 1 metre from 1990 levels, depending on how much greenhouse gas we continue to add to the atmosphere (and there’s an excellent set of slides explaining those predictions here). But scientists are still trying to understand how ice sheets, particularly in the Antarctic, could contribute to sea level rise.  A recent study suggests that Antarctic ice sheets alone could drive a metre of sea level rise by 2100 (DeConto & Pollard, 2016), while others scientists are examining whether such rapid changes are physically possible (Ritz et al., 2015).

More than 300 million people live in coastal areas already being affected by or at risk of impacts from predicted sea level rise, including 130 coastal cities each with more than 1 million inhabitants.  But sea level rise isn’t the same everywhere, like adding water to a bathtub – instead, its effects will vary on different coasts, for example as water sloshes around the oceans through currents and local variations in gravity.  A new analysis shows which coasts are connected to changes in particular ice sheets – with changes in the Antarctic Peninsula being particularly relevant for Sydney, Australia (Larour et al., 2017; and here’s an interactive website from that research that allows you to explore how different cities are affected by sea level changes from ice in different areas).

Unless we adapt for predicted sea level rise with new defences and relocations, the financial costs in damages from storm surges and coastal flooding each year could total nearly ten percent of the world’s Gross Domestic Product by 2100 (Hinkel et al., 2014).  Even if you don’t live near the coast personally, you will be affected by sea level rise through the global economy. So although the processes involved are complex, what’s happening right now in the Antarctic – along with changes in other places where ice drains from land into the ocean – affects us all.

I’d like to thank the BBC Natural History Unit’s Blue Planet II team for the opportunity to join their Antarctic expedition, and the ship’s crew, sub team, and expedition members aboard the  Alucia, whose hard work gave us a new view of life on the ocean floor around Antarctica, and the chance to collect some more data to help scientists understand how the environment is changing there.

There are days when I love my job, & today was one of them, thanks to the hard work of the whole team aboard the research ship #Alucia pic.twitter.com/mTb5Mk9Pjr

— Dr Jon Copley (@expeditionlog) December 14, 2016