Fissures and steep terraces along the leading edge of Thwaites glacier are causing it to melt faster, whereas pooling of cool water is slowing melting at the glacier's base, new research has found.
Thwaites is one of Antarctica's least stable glaciers, and this information helps researchers accurately model how quickly it's retreating.
"Our research certainly points to Thwaites melting in a different manner than was expected," study co-author Peter Washam from Cornell University said.
The glacier holds enough water to raise global sea levels by 65 centimetres, and has retreated by about 14 kilometres since the late 1990s.
It also acts as a buffer, holding back neighbouring glaciers that contain around three metres of potential sea level rise.
Some scientists fear Thwaites may have already reached an irreversible tipping point, where it will be completely lost regardless of future greenhouse gas emissions pathways.
However, even if that's the case, there's a lot of uncertainty around how quickly it will melt.
That's because there are so many interacting forces involved with glacier retreat, and research is difficult to conduct in the extreme Antarctic environment.
Two studies published today in Nature used a new underwater vehicle to study the sea floor and waters around the Thwaites Eastern Ice Shelf, shedding new light on what's happening deep beneath the ice.
The researchers deployed the sophisticated vehicle called "Icefin" down a borehole, made using a hot-water drill through hundreds of metres of glacier, said glaciologist Ben Galton-Fenzi from the Australian Antarctic Division, who wasn't involved with the research.
"[Icefin] unfolds itself as it gets into the ocean under the ice," Dr Galton-Fenzi said.
Using Icefin, they took a series of measurements including temperature, dissolved oxygen and salinity, and mapped and imaged the glacier and sea floor around the grounding line — the zone where the glacier lifts from the sea floor and begins to float at around 500m deep.
Glacier thinning increases six-fold
What they found is that average water temperatures underneath the ice shelf are "substantially warmer" than melting point. However, water temperatures aren't evenly distributed.
Where there was low current, the researchers found that cooler, fresher meltwater settled out from the saltier sea water into gradients, a process known as stratification.
But there were broadly two variables at odds with one another. First, cooler water is typically denser than warm water, and so should be heavier. And second, fresh water is lighter than saltier water.
In the case of cool, fresh water, the salt content was the determining factor — cool, fresh water was lighter and tended to rise toward the surface.
In places where the topography or shape of the ice shelf — the part extending out from the grounding line — was fairly flat, they found cool, fresh water would pool under the ice, buffering its rate of melting.
In these areas, the researchers estimated that the glacier is thinning by an average of around 5 metres per year.
But where the topography of the ice shelf rose steeply, or where crevasses were present, water current and mixing were significantly higher as the fresh water pushed upward.
This mixing brought the warmer salt water in contact with the ice, producing melting and thinning rates of as much as 30m per year, or more in some extreme cases.
"When you've got a bit of melting under the steep slope, the freshening makes the ocean a lot lighter," Dr Galton-Fenzi said.
"That wants to start the flow up the slope — we call it 'buoyancy driven circulation'. It wants to melt more and more and more, and you get this positive [melting] feedback."
Implications for global sea level
The new data will be invaluable in improving modelling not just of Thwaites glacier retreat, but glaciers more generally, according to Poul Christoffersen, a glaciologist from the University of Cambridge who wasn't involved with the research.
"They're really difficult observations to make. It's great new evidence from one of the most important glaciological sites — Thwaites glacier being one of Antarctica's weak points," Professor Christoffersen said.
"Undoubtedly this will help improve model parameters and behaviours."
But it's not just Thwaites that we need to be watching, according to Dr Galton-Fenzi.
"The two in [Australia's] backyard are the Aurora sub-glacial basin and the Wilkes sub-glacial basin.
"They have the potential to raise sea level by up to 12 metres."
According to Dr Galton-Fenzi, funding cuts and funding redirection mean Australian researchers have lost much of their capacity to monitor glacier retreat in Antarctica.
"The [Australian Antarctic Division] has been incredibly under-resourced. We've lost our hot-water drilling capacity."
However, he said they're in the process of rebuilding.
"There's a new initiative called REACT — Risk of East Antarctica Collapse and Tipping-points — that's designed to develop this capability we need to understand the ice sheets.
"The plan is within five years we'll have access to those areas in East Antarctica that are changing and vulnerable, and to [re-establish] our own hot-water drilling capability."
Dr Galton-Fenzi said this latest research — and the tools used — can play an important role in helping us better understand the threat of sea level rise.
"The big things we need to look at are: what's the risk of Antarctic ice sheet collapse, and what are the implications for global sea level?"
