Tag Archives: Research

Nature: Heterogeneous melting near the Thwaites Glacier grounding line

In late 2019 and early 2020, the MELT team took Icefin beneath Thwaites Glacier (sometimes unfortunately referred to as the Doomsday Glacier). Thwaites is the size of the U.S. state of Florida (or about the size of Great Britain), and is one of the most rapidly changing glaciers on Antarctica. Using a 600 m borehole drilled through the ice by the British Antarctic Survey, members of the MELT team were able to deploy Icefin and other sensors beneath the glacier and maneuver it all the way up to the point where the underside of the ice first detaches from the seafloor, known as the grounding line – a point which was previously unseen, until Icefin. By doing this, the MELT team of scientists and engineers are uncovering critical information about the mechanisms that influence how fast this massive glacier is melting. The team chronicled their findings in two Nature papers, led by the Icefin team and the British Antarctic Survey (Schmidt et al. 2023 and Davis et al., 2023), and here’s what they found.

Figure 1 from Schmidt et al., 2023, Nature 614:7948, https://doi.org/10.1038/s41586-022-05691-0

The Grounding Line

Figure 1 shows the ocean cavity beneath Thwaites Glacier, sampled with Icefin, leading up to the grounding line, which is the final contact point between the ice and seafloor before the ice goes afloat to form an ice shelf.  All of the data in panels b & c come from Icefin sensors: the ice shape (gray) comes from Icefin’s altimeter, sea floor from sonar & DVL, temperature from our CTD.

The team found that the ocean was warmest and saltiest deep, near the seafloor, but cooled and freshened near the ice base due to melting (here the temperatures are given in thermal driving, which is temperature above the water’s freezing point at that depth). Colder ocean conditions were found very close to the grounding line. However, the ocean remained well above freezing throughout most of the ocean cavity, and therefore contained ample heat to melt the ice. The presence of such a warm ocean beneath Thwaites means that it is capable of melting the ice rapidly.

Figure 2 from Schmidt et al., 2023, Nature 614:7948, https://doi.org/10.1038/s41586-022-05691-0

Grounding Line history left in the seafloor

Figure 2 shows features carved into the seafloor by the ice base of Thwaites Glacier as it flowed over it, when the ice was grounded further towards the ocean. The shape, size, and orientation of these features tells us the history of the region as the grounding line retreated across it.  This data comes from our Norbit mapping sonar and Oculus Forward sonar.

The team found that the seafloor in the survey region contained ridges and troughs that were oriented in the direction that Thwaites Glacier flows. These ridges and troughs extended all the way to the grounding line, where their shape is mirrored by the ice base. A single wedge of sediment cut across these streaking features, from a time when the grounding line stuck in place, called a “grounding line wedge”. Except for this one occasion, there were no other grounding line wedges, meaning that the grounding line retreated steadily backwards across this region over the whole visible period (about the last 10 years). A possible gully was also present in the seafloor near the grounding line, where the team thinks that meltwater may have flowed out like a river from beneath the glacier upstream into the ocean. 

Up close and personal with the ice

Figure 3 shows that the ocean conditions are highly variable right up to the underside of Thwaites Glacier. Various factors change depending on distance from the grounding line and whether below flat  or steeply sloped ice.  Here, data from the forward and upward cameras and the CTD and JFE Rinko Oxygen sensor show that melt water collects at the top of features in flat spaces, and very warm water melts steep slopes.

The team found that the ocean conditions and the amount of meltwater near the ice base changed depending on the shape of the ice base. Beneath steeply sloping parts of the ice the water remained warm and the ice contained scallop features from strong melting. Beneath flat parts of the ice the water was considerably cooler and fresher. The ice surface was featureless here, because the cold and fresh meltwater inhibited the warm ocean from melting the ice.

This figure shows how ocean conditions influence ice base morphology, which varies with distance from the GL (a through c).
Figure 3 from Schmidt et al., 2023, Nature 614:7948, https://doi.org/10.1038/s41586-022-05691-0
Figure 4 from Schmidt et al., 2023, Nature 614:7948, https://doi.org/10.1038/s41586-022-05691-0

Ocean flow in crevasses and terraces

Figure 4 shows how fast the ocean flows close to the steep sides of terraces and crevasses.

The team found that ocean flow speeds increased in crevasses, which helped to mix heat and salt into their steep sides and melt them. The ocean flow decreased beneath the flat ice base, due to friction, which sheltered any terraces above it.

Not all melting is created equal

Figure 5 shows how different ocean conditions influence the way the underside of Thwaites Glacier melts. Various factors, which change depending on the distance from the grounding line, cause the ice to melt at different speeds.

The team found that, while in some areas along the base of the glacier the ice is melting slower than expected, in other areas like steeply sloped terrace walls and crevasses, the ice is melting much more quickly than previously known. This rapid melting in sloped areas of the underside of the glacier may contribute to the ongoing retreat and eventual collapse of Thwaites, which in turn will have a significant effect on sea level rise.  In fact, warm waters in the crevasses are making these cracks wider, contributing to the eventual break up of the ice.

Figure 5 from Schmidt et al., 2023, Nature 614:7948, https://doi.org/10.1038/s41586-022-05691-0
Figure 6 from Schmidt et al., 2023, Nature 614:7948, https://doi.org/10.1038/s41586-022-05691-0

The slope-melt feedback

Figure 6 further shows how the ice melts at different speeds, depending on the slope of the ice, and how important this is to the overall melting of the glacier.

The team found that the melt rate increased with the sine of the ice base slope. This led to strong melting of 30 m yr-1 when the ice was vertical and melting of around 5 m yr-1 where the ice was completely flat. Overall, they found that 27% of the total melting occurred at slopes greater than 30°, despite this being a relatively small fraction of the total ice base (about 9%). This means that any part of the ice with a steep slope is very important to the overall melting of the ice.

More images of the ice base melting

This series of nine images in Extended Data Figure 8 from the paper shows examples of the strange topography and strongly asymmetric melting under Thwaites Glacier in Antarctica, photographed by Icefin.

The images from Icefin show conclusively that the ice base beneath different parts of the glacier melts in different manners. Scallop formations on terrace walls and sediment raining down from ice ridges show that stronger melting occurs here than the flat and featureless ice base. These fascinating images will stimulate further investigations into how the ice and ocean interact in Antarctica. 

This series of nine images shows strongly asymmetric melting and melt processes under Thwaites Glacier in Antarctica, as explored and photographed by Icefin.
Extended Data Figure 8 from Schmidt et al., 2023, Nature 614:7948, https://doi.org/10.1038/s41586-022-05691-0

Thwaites MELT is a team effort!! Our colleagues at the British Antarctic Survey, led by Peter Davis, also published a companion paper about results from the oceanographic data and the mooring left behind at Thwaites. Please see this other paper here.

None of this work would have been possible without the efforts of Professor David Vaughan of the British Antarctic Survey, coauthor on the Icefin paper, who sadly left us in February, 2023. A great scientist and wonderful friend and mentor, he will be sorely missed.

First look under Thwaites Glacier and Kamb Ice Stream

Georgia Tech scientists get first look deep under Antarctica’s Thwaites Glacier and Kamb Ice Stream

ANTARCTICA — An international team including scientists from Georgia Tech captured new images and first-of-its-kind data from deep beneath an Antarctic glacier, which will help scientists to better understand the impact of one of Antarctica’s fastest changing regions and its impact on future sea level rise.

"It's our 'walking on the moon' moment." Thwaites Glacier Icefin quote

Their work will be featured as part of a special report on BBC World News on Tuesday, Jan. 28, in celebration of the 200th anniversary of the discovery of Antarctica.

Stationed in Antarctica for the last two months, the MELT (Melting at Thwaites grounding zone and its control on sea level) team, part of the International Thwaites Glacier Collaboration, deployed ocean instruments and cored sediments to gather data on one of the most important and hazardous glaciers in Antarctica. The MELT team included Georgia Tech scientists who used an underwater robot named Icefin to navigate the waters beneath Thwaites Glacier and collect data from the grounding zone – the area where the glacier meets the sea.

Dr. Britney Schmidt, lead scientist for Icefin and associate professor in Georgia Tech’s School of Earth and Atmospheric Sciences, said the new data represented several firsts for her team, as well as for science as a whole.

“We designed Icefin to be able to finally enable access to grounding zones of glaciers, places where observations have been nearly impossible, but where rapid change is taking place,” said Schmidt, a co-investigator on the MELT project. “We’re proud of Icefin, since it represents a new way of looking at glaciers and ice shelves. For really the first time, we can drive miles under the ice to measure and map processes we can’t otherwise reach. We’ve taken the first close-up look at a grounding zone. It’s our ‘walking on the moon’ moment.”


Located in a remote part of Antarctica, where few scientists have ever ventured, the team battled sometimes hostile weather, extreme winds, and temperatures below -22 degrees Fahrenheit to get close enough to the Antarctic coastline for Icefin to reach the grounding zone.

In these trying conditions, the MELT team used hot water to drill through up to 2,300 feet – nearly a half mile – of ice to get to the ocean and the seafloor below. On Jan. 9 and 10, Icefin swam more than a mile from the drill site to the Thwaites grounding zone, to measure, image, and map the glacier’s melting and gather other important data that scientists can use to understand the changing landscape and conditions. Not only did the team put one Icefin robot down the borehole at Thwaites Glacier, but they did it with a second Icefin vehicle in collaboration with Antarctica New Zealand near the grounding zone of Kamb Ice Stream, part of the Ross Ice Shelf.

Thwaites Glacier, which covers an area the size of Florida, is particularly susceptible to climate and ocean changes. Thwaites melting accounts for about 4 percent of global sea level rise, and the amount of ice flowing out of Thwaites and its neighbouring glaciers has nearly doubled in the past 30 years, making it one of Antarctica’s most rapidly changing regions.

"We're particularly concerned about Thwaites." Thwaites Glacier quoteDr. Keith Nicholls, an oceanographer from British Antarctic Survey and UK lead on the MELT team, said Icefin’s exploration of sediment and other conditions in the Thwaites grounding zone will help scientists determine how this region will change in the future and what kind of impact on sea level rise we can expect from these changes. The MELT team also deployed radars and oceanographic sensors, conducted seismic studies and took sediment cores from beneath the glacier, and deployed two moorings through the ice that will record ocean and ice conditions for the coming year to monitor changes at Thwaites.

“We know that warmer ocean waters are eroding many of West Antarctica’s glaciers, but we’re particularly concerned about Thwaites,” he said. “This new data will provide a new perspective of the processes taking place, so we can predict future change with more certainty”

The MELT project is funded by the International Thwaites Glacier Collaboration (ITGC), a collaboration between the U.S.’s National Science Foundation and the UK’s Natural Environment Research Council.

From left, (1) Icefin image of sediment laden ice at the grounding zone of Thwaites Glacier, Antarctica. (2) Icefin view of the grounding zone of Thwaites Glacier, Antarctica, in less than one meter of water. (3) Icefin image of sediments and rock in the ice at the grounding zone of Thwaites Glacier, Antarctica.
From left, (1) Icefin image of sediment laden ice at the grounding zone of Thwaites Glacier, Antarctica. (2) Icefin view of the grounding zone of Thwaites Glacier, Antarctica, in less than one meter of water. (3) Icefin image of sediments and rock in the ice at the grounding zone of Thwaites Glacier, Antarctica.

“To have the chance to do this at Thwaites Glacier, which is such a critical hinge point in West Antarctica, is a dream come true for me and my team. The data couldn’t be more exciting,” Schmidt said. “And exploring the grounding zones of two different glaciers in the same season is incredible.”

In addition to the MELT project, Schmidt is the Primary Investigator for the RISE UP (Ross Ice Shelf and Europa Underwater Probe) project, which also had team members from Georgia Tech deployed in Antarctica this season.  RISE UP is a NASA-funded project that developed Icefin from a prototype to a full-fledged underwater vehicle and aims to develop technology for future missions to Jupiter’s moon Europa.

Both the MELT and RISE UP teams spent time at McMurdo Station, Antarctica conducting research, before simultaneously deploying to more remote areas. Antarctic logistics for both projects were supported by the National Science Foundation, under the United States Antarctic Program.

RISE UP‘s work at Kamb Ice Stream came as part of a collaboration with two projects supported by Antarctica New Zealand: the NZARI Ross Ice Shelf Programme led by Dr Christina Hulbe of the University of Otago, and the NZ Antarctic Science Platform’s Antarctic Ice Dynamics project, led by Dr Huw Horgan of Victoria University. 

RISE UP team members deployed along with the New Zealand hot water drilling and science teams to study the Kamb Ice Stream – a river of ice – on the Ross Ice Shelf in Antarctica. Their goal was to explore and map areas near the grounding zone to better understand its flow and the surrounding environment.  Icefin’s work at Kamb Ice Stream will continue next season as part of Dr. Horgan’s project.

“We now have, effectively, a transect of conditions from the front of the Ross Ice Shelf to the grounding line,” sadi Christina Hulbe of the Ross Ice Shelf Programme, which finished its final year of field work in late December. “In addition to Icefin’s work, we’ve installed our third ice-anchored mooring, collected cores for sedimentary and microbiological analysis, we’ve imaged the ice optically and using radar, and made high resolution observations of ocean conditions.”

The RISE UP team completed three dives with Icefin, and team member Ben Hurwitz, a graduate student at Georgia Tech who works on Icefin’s technology, said the season was wildly successful, adding the team was “excited to share what we found in the coming months.”

Notes on the projects:

The International Thwaites Glacier Collaboration: www.thwaitesglacier.org

The MELT Project is lead by Keith Nicholls , an oceanographer with the British Antarctic Survey (BAS), and Dr. David Holland, an applied mathematician (with a background in fluid dynamics) at New York University, with co-leads Dr. Eric Rignot from the University of California at Irving, Dr. John Paden with George Mason University, Dr. Sridhar Anandakrishnan out of Pennsylvania State University, and Dr. Britney Schmidt at the Georgia Institute of Technology.

RISE UP’s field work at Kamb Ice Stream came as part of two science projects funded by Antarctica New Zealand and the Victoria University of Wellington Science Drilling Office. The other research partners involved on the project are: The University of Otago, Victoria University, University of Canterbury and University of Waikato, NIWA and GNS Science from NZ and the ROSETTA project and Universty of California, Santa Cruz in the US.

On the eve of deployment

(Note: This post was written before a week-long delay due to weather and before Matt, our lead engineer, managed to make it out, but most of it still holds.)

I’m anxious.

Not nervous. Anxious.

Our bags are packed. The vehicle and it’s associated spares, tools, and other miscellany have been boxed and are sitting in the airfield waiting to be loaded into a Basler tomorrow for our deep field deployment to the Kamb Ice Stream.

The anxiety comes from the waiting.

Three years of waiting for this to happen.

The deep field.

Maybe that should read “The Deep Field”.

It feels like something that’s so far away, both spatially and temporally. But within twelve hours, we’ll be heading there to do something that nobody has ever done before.

Three years of preparing.

More for Britney. I can’t imagine her anxiety levels. And she won’t even be there when the vehicle accomplishes its mission, its purpose, for the first time.

(Don’t worry, we’ll take pictures.)

660 meters of ice. That’s 2165 feet. Half a mile. With a 14″ wide hole down, down, down to the bottom. To the underside of the Ross Ice Shelf.

The Ross!

The largest ice shelf in the world. (By area only, as Peter keeps saying; by volume, the Filchner-Ronne is larger).

And we’re going to the bottom.

The anxiety is real.

Does that come across in this blog? Are you anxious with me?

I’ve never been to The Deep Field before. I hope I packed right. Mostly that’s base layers, socks, and a toothbrush.

A month in the middle of nowhere.

Never before will I have been so far from, well, anything.

KIS-1 (“Kamb ice stream 1”), also called HWD-1 (“Hot water drill 1”), is located at 82.77S, 156.57E. The nearest anything is Simple Dome camp, at 81.65417S, 149.005E. And that’s a field camp, not exactly a bustling place. But that would be the closest actual place, 105 miles approximately north.

Select deep-field deployment locations for USAP and ANZ
The Ross Ice Shelf and surrounding region with select USAP and ANZ deep field sites marked. McMurdo and Scott Base locations are marked separately.

That’s an 11 minute flight, for those counting at home, according to WolframAlpha. Or 760 microseconds, if you’re a photon.

Looking at it from that perspective, it’s possible I’ve been farther from people before.

But I don’t think that it’ll feel that way once we’re there.

Due to some unforeseen circumstances (the C-17 delays and the weather at other sites delaying flights), we’ve gotten lucky. We’ll be flying out a touch earlier than expected, and we’ll be flying with the drillers. So we’ll be among the first out there. This is extremely valuable; the Thwaites project has been massively behind due to the same flight delays, and thus all flights from McMurdo have been pointed towards their staging ground at WAIS Divide (79.4675 S, 112.0864 E). Thus we need to get a flight to KIS-1 before the Thwaites flights start taking people. Because then we become a much lower priority for USAP. Which is fair; Thwaites is a $25 million project, whereas KIS-1 is not even a US-sponsored field camp.

The Basler we’ll be flying is a modified DC-3. A DC-3! The plane was around prior to WWII! The Baslers, though, have excellent records in Antarctica. They have upgraded avionics and engines, modified fuselage and wings, and a host of other new features. Which is reassuring. I was a little perturbed when I was told it was a DC-3.

By the time you read this, we’re likely to already be in camp, setting up tents, building snow walls to ward off the wind, and checking out the vehicle. I’ve been told that The Deep Field focuses you. You’re forced to have singular purpose. Distractions are removed from sight and mind.

But man, am I anxious.