Five Days of Work for 5000 Years of History: A Closer Look at Shallow Ice Core Drilling

April 3, 2020, 5:43 PM EDT

Above: A core section of ice from more than 500 feet below the surface of Antarctica sees the light of day for the first time in over 3000 years. The ice drilling rig is seen at left, in operation at the PALEO site in East Antarctica. (All photos by Pete Akers unless otherwise noted.)

Editor's note: Paleoclimate researcher Pete Akers (Institut des Géosciences de l’Enivironnement or IGE in Grenoble, France) is a participant in the 2019-20 East Antarctic International Ice Sheet Traverse (Project EAIIST). Pete has been writing about the project in a special series for Category 6. See the bottom of this post for links to prior posts.

With the Antarctic field work of the EAIIST project wrapped up over a month ago and all field members back at their labs across Europe, the main focus of the EAIIST team recently has been to get back into a sense of normalcy at work. For some members whose data was collected directly in the field (such as the ice-penetrating radar and snow physics observations), analysis could begin immediately upon their return. For those needing to perform chemical analysis on collected snow samples, such as me, we have a bit of down time to focus on other projects until the samples arrive in France sometime in June.

Like many others worldwide, our lab has not escaped the impact of COVID-19. With the government of France ordering non-essential working people to remain at home for at least two weeks, CNRS closed the IGE laboratory and offices. While we have cold rooms to store and analyze snow and ice at IGE, these have limited back-up capabilities and the decision was made to shut them down while the lab is closed. Samples currently held in our cold rooms were taken back to the main storage facility housed within a frozen-food warehouse nearby until the lab can reopen. In the meantime, IGE and other CNRS research groups have collected lab supplies that are in medical shortage, such as gloves and alcohol, and given them to the local hospital system in Grenoble.

With our research fully shifted to remote working and manuscript writing, this seemed a perfect opportunity to wrap up my main series on the EAIIST project with a closer look at ice coring in the field.

While several shallow snow and ice cores were taken at different sites along the EAIIST traverse, we drilled our longest core at a site called PALEO, located 90 miles (150 km) north of the Megadunes and 210 miles (500 km) south of Concordia Station. This site was chosen because it is drier and colder than Concordia, where the EPICA Dome C core was drilled. Although the Megadunes region is even drier and colder than PALEO, the dunes themselves slowly migrate across the landscape over the years (like wind blowing sand dunes) and this makes interpreting an ice core more difficult. For example, it can be hard to tell if a chemical or physical change in the ice was due to climate change or simply because a dune passed over at that point in the past.

Coring for the mile-deep ice cores at sites like Dome C and Vostok are multi-year efforts, with impressive mechanical innovations and housing infrastructure to support the project. In contrast, field drilling of shallow cores (i.e., cores less than 1000 ft long) is a relatively simple endeavor. Several different models of ice drilling rigs exist with different capabilities, core diameters, and maximum depths. Determining which model to use depends largely on how much ice you need for your analysis. Typically, a single core will be cut lengthwise into several narrower pieces for multiple chemical analyses. A larger-diameter ice core gives you more ice to analyze for each year of accumulation, but increases the weight and number of boxes needed to haul and ship back.

On our traverse, we brought three different ice drilling rig models, but because of problems with two of them, we ended up drilling all the needed cores with the large Trapanelle model, which can drill down to 200 m and makes cores with a 4-inch (10-cm) diameter.

Although the PALEO location was specifically chosen for a core for the reasons listed above, the exact location of the core at the general PALEO site wasn’t important. When coring a lake or even a mountain glacier, more focus must be made to choose a specific coring spot with the best likelihood for high quality and long record. Here on the ice sheet, the snow and ice are pretty much all the same anywhere for miles in any direction, so we simply set up the drilling rig at a spot a dozen yards upwind of where we parked our caravan and got to work.

Drilling a shallow ice core

We hoisted the Trapanelle off the side of the caravan where it had been stored with a crane attached to one of our driving tractors and placed it at our chosen drilling spot. Power is supplied from the main caravan generator, and we erected three large tarps upwind to give us some blocking of the chilly winds. Temperatures were relatively mild most days—between 15 and -22°F (-25 and -30°C)—with two windier and cold afternoons and one extraordinarily calm and sunny afternoon that approached 0°F (-17°C).

Work generally started after breakfast around 8 am and continued until dinner after 6 pm, with some breaks for coffee and lunch. In our full polar gear and wind blocking, we mostly stayed plenty warm (and even got a little too hot on the extra-warm day!). And at these temperatures, there is no danger of the ice melting, even in direct sunlight, so all work is simply done outside.

To start an ice core, the drilling piece of the Trapanelle is raised to vertical and a winch lowers it until the drilling bit makes contact with the snow surface. The drilling operator then activates the drill motor through the control panel. This drill motor is located directly above the drill bit, and it spins the bit to carve into the snow. The bit will carve down until it reaches its maximum depth of 3 ft (1 m), at which point the drill motor is turned off and the winch reversed to retrieve the drilling piece. Spring-loaded backward facing prongs at the bottom of the drill bit serve as a “core-catcher” and keep the newly drilled core from falling out the bottom during retrieval.

After the first core is taken, subsequent rounds of drilling are guided into the same borehole and the drilling piece is lowered down to the last cored depth to drill another 3 ft (1 m). Three metal pieces above the drill motor press against the walls of the borehole and keep the drill bit from wildly spinning or bouncing as it descends and ascends. The “snow” created as the bit cuts down into the ice is gathered and held under the protective bit cover to keep it from piling up in the borehole. Once the drill is back at the surface, the drill motor is reversed to remove this “snow” trapped between the protective casing and the drill bit.

The drilling piece is then moved to horizontal and the drill bit detached from the larger piece. The 3-ft (1-m) drill bit is removed, with the core still inside, and moved over to a wooden support structure. The top of the bit (and core) are pointed down, and a plastic plug that kept the drilling-produced “snow” away from the core is removed.

With one team member standing at the drill bit top, another member raises the bottom of the drill bit so that the ice core inside slides out the top. The ice core is then moved over to a nearby table and placed in a V-shaped aluminum trough. It is very important that the top of the ice core is placed the left in the trough, as there is no way to tell the top of a core from the bottom once it is out of the bit. Instead, we make sure that the top of the core is always facing to the left through the entire processing period.

As two team members place the drill bit back in the rig and start the process of drilling the next core, another member begins to measure and process the ice core just drilled. If we are lucky, the core will be in one piece and close to the desired 3-ft (1-m) length. However, the actual core length varies based on where exactly the bottom of the core broke from the main ice mass when pulling the drill bit back to the surface. The Styrofoam boxes we use to store and transport the ice cores are only 3 ft (1 m) long inside, so cores are packaged at this length for maximum efficiency. If a core is too long, we saw the excess off and simply add it to the top of the next core. If a core is shorter than 3 ft (1 m), we will wait and take a section of the next drilled core to get it up to the desired length.

After the length of the core is measured and logged, we bag the core in clean plastic with the top and bottom of the core clearly labeled along with the core ID. After bagging, the core is placed in the Styrofoam storage box ready for transport.

The tempo of an ice coring session

At the beginning of drilling a core, everyone on the team (usually three people) is in constant motion, with cores rapidly being drilled and delivered to the processing table. I usually worked as the table processor, measuring and cutting cores to length and also preparing the bags and Styrofoam boxes. The actual drilling of the core and extracting it from the bit only takes 5-10 minutes, and processing takes a little longer than that, so for the first 20 cores, I was mainly trying to avoid getting too backlogged.

However, the longest part of the drilling process soon became apparent. After 60 ft (20 m) or so, it takes 5-10 minutes simply to raise and lower the drilling bit, and the pace of work becomes more relaxed. While the bit is being lowered or raised, there’s not much to do other than keep an eye on the bit depth. After 100 m of drilling, the wait between cores grows to over 20 minutes, with a quick rush of activity upon core retrieval followed by another stretch of waiting.

Toward the end of our coring, we could get only two or three cores per hour, and cores would increasingly come up shorter than a meter and broken as the ice was more brittle at depth. Sometimes the core came up in several fractured pieces, and it became a bit of a puzzle to figure out their orientation and order. By the time we got to a depth of 590 ft (180 m), the cores were breaking too much for us to make sense of the pieces and the length of time to core was impractical.

As we shut down the operation after five straight days of coring, we celebrated the end of our full core, which covers over 5000 years of snowfall and climate history.

Above: A timelapse taken over a few hours at PALEO when our team was drilling at depths below 400 ft (120 m). The wooden structure at left is used to support the drill bit when removing the core, and the table at right is used for processing. At this point in drilling, there’s about 20 minutes of down time between cores. I am the one processing the cores on the table. Other things to look for besides the core recovery process: me changing our music selection through my phone deeply buried in my polar gear, the other guys taking walks to pass the time, me filming parts of the process for our documentary, and how I shoveled snow into the Styrofoam container to serve as packing material.

Changing from snow to ice

As mentioned in a previous post, snow that falls on the surface of an ice sheet becomes glacial ice over time as it gets buried and subject to higher and higher pressure. One of the highlights for me in helping to drill the entire 200-meter-long ice core at PALEO was seeing the changes in the individual cores with depth. In the top 10-20 feet (3-6 m), the cores are still largely snow-like and have a consistency similar to a packed snowball. They are quite fragile and often break into several pieces during retrieval and processing, so extra care is made to keep them in the correct order and orientation. With the limited snowfall at PALEO—less than 4 in (10 cm) per year on average—each core near the surface represents 10 or more years of snow accumulation.

As you drill below this top snow zone, the individual grains of snow get larger as the pressure consolidates them into little bits of ice. You end up with cores that look a little like a snow cone in consistency and density, but they generally hold their core shape well. This stage, intermediate between snow and solid ice, is known as “firn”.

By around 300 ft (100 m) in depth, the increased pressure from the weight of the overlying snow and firn has fused the icy grains of firn into a solid mass of ice. Each individual core is now much heavier as well due to the increased density, and this depth has ice that is over 2000 years old (based on dated cores from Dome C). However, the air that was originally in the gaps between the icy grains of firn becomes trapped as bubbles within the ice. These bubbles make the ice still largely opaque as a core, but individual flakes or pieces of the ice here can be quite beautiful in the sun. These bubbles are also how we can use ice cores to reconstruct past atmospheric compositions: once the air gets trapped in the bubbles, it is preserved until we melt or crack the ice and release it again.

As you keep drilling deeper below 500 ft (150 m), the bubbles in the ice and bubbles get compressed more and more (the bubbles here are at ~200 psi), and the cores become denser and more translucent, but also quite brittle. They are also very cold, as the temperature within the ice sheet here is near -60°F (-55°C). Luckily, there’s not much handling of the cores, although I did get a few spots of frost nip on my fingertips from gripping the cores during processing, despite wearing gloves.

I did, however, discover a layer of volcanic ash in one core around 400 ft (120 m) down. Though the layer was very faint, anything in the ice that isn’t bubbles immediately sticks out when you’ve been staring at pure ice cores for several days. The ash probably came from an Antarctic volcano to the west in the Transantarctic Range and fell around 3000 years ago. Comparing this layer with known volcanic layers in other Antarctic ice cores may help us better date our PALEO core.

Half of our PALEO ice core will go to Australian colleagues who will focus their analysis on cosmogenic nuclides, rare isotopes of elements that are created when cosmic rays interact with material on Earth such as the atmosphere. By tracking how these nuclides change in concentration over time, we can learn things such as how solar activity varied in the past (because a stronger sun means a stronger solar wind that affects cosmic ray penetration on Earth). The rest of the core will go to Italian and French researchers for more-standard ice core analysis, looking at water isotopes and trapped aerosols.

Currently, the boxes of ice cores from PALEO are being held in a freezer shipping container in Australia, where some initial cutting and processing began to supply the Australians with their samples. Eventually, this container will be shipped back to France and make its way to Grenoble, although the COVID-19 pandemic may alter the timing of this shipping. Full analysis of the PALEO core and other samples will take over a year, with published results likely not for another year as well. In the meantime, I will do my best to provide intermittent updates as I can!

As this post wraps up this series on EAIIST and my Antarctic research, I’d like to thank everyone for following along the journey. Please feel free to contact me through emails or my Twitter account, @petescientist (where I largely post Antarctic research photos and updates) if you have any questions in the coming months. I hope you’ve enjoyed the posts and have a bit better idea now of what happens in the field of polar science.


Previous posts in this series:

Exploring East Antarctica and Its Role in Climate: A First-Hand Report (November 7)

A Frozen Realm: Inside the East Antarctic Ice Sheet (November 21)

Arrival in Antarctica: Not Your Everyday Trip (November 29)

A Taste of the Traverse: How to Cross Antarctica without a Plane (December 26)

Life and Work on the Frozen Continent: Antarctic Research Stations (January 17)

Snow, Stars, and Stress: Science at Concordia Station (January 24)

Where None Have Stood Before: A Return to the EAIIST Traverse (February 19)

The views of the author are his/her own and do not necessarily represent the position of The Weather Company or its parent, IBM.

author image

Pete Akers

Pete Akers is a postdoctoral researcher with CNRS in Grenoble, France, where he studies the chemistry of ice cores to reconstruct past climate changes. His previous paleoclimate research has brought him to Maya ruins, Indiana caves, and the Greenland Ice Sheet.

Recent Articles


That Was Fast: Tropical Storm Bertha Develops and Makes Landfall Over South Carolina

Bob Henson

Section: Hurricanes, Typhoons & Cyclones


Slight Chance of Southeast Development from 91L; Watching Eastern Pacific for a Weekend System

Bob Henson

Section: Hurricanes, Typhoons & Cyclones


U.S. Hailstone and Hailstorm Records

Christopher C. Burt

Section: Miscellaneous

Please note that DISQUS operates this forum. When you sign in to comment, your sign in information, along with your comments, will be governed by DISQUS' privacy policy. By commenting, you are accepting the DISQUS terms of service.

The comments made below do not necessarily represent the views of Weather Underground; The Weather Company, an IBM Business; or IBM. Comments below should not be perceived as official forecasts or emergency information. For official information on potential storm impacts and evacuation information, please follow guidance from your local authority's emergency operations department.