We found water!

January 3, 2019

One of our main science objectives was to confirm the existence and persistence of meltwater in the firn at two ice shelf sites. For the Wilkins Ice Shelf, the remote sensing microwave data combined with airborne radar profiles showed very clear evidence of water. For the George VI Ice Shelf, the available data was not as convincing—at best, we hoped for a few thin layers of wet firn. To examine the snow and firn layering first-hand, we used a lightweight thermal ring drill specifically made for extracting narrow cores of ice and snow at the melting point. This is more accurate than using a dousing rod.

In a nutshell, the thermal drill is an electrical heat ring attached at the end of a long narrow metal barrel. The drill melts an annulus of firn or ice around the extracted core, which slowly rides up into the barrel as the hot ring progresses through the snow, one meter at a time. Core dogs, or small sharp metal teeth just above the heat ring, bite into the core and break it off as the drill is pulled back up to the surface. The barrel is attached to a Kevlar-reinforced cable that includes electric power wires for the heat ring. At the surface, the cable goes through a wheel on top of a tripod to facilitate raising and lowering the drill and keeping it plumb.

The morning after our arrival at the Wilkins Ice Shelf site, we set up the drill and started working. In the photos below, you can see the different steps we took for coring and processing the core. After drilling for a meter in the borehole and breaking the core free from the bottom, we pulled the drill barrel back to the surface and laid it on its side, using magnets to prevent the metal core dogs from catching the core. As we tilted the barrel, the core slide out on to a measurement table, a board for photographing and cutting the core. Once the core was on the tray table, it was cut into ~20 centimeter sections. Each section was measured for its length and diameter, and then weighed. We also described the stratigraphy and measured the thickness of each ice layer. For the very top part of the core (in our case, the upper meter and a half), the density was difficult to measure accurately because the snow core was fragile enough to compact during the drilling process. We dug a snowpit next to the borehole and measured the density of this uppermost section using a snow-density kit.

On our first drilling day, we reached about 10 meters depth and were still waiting for our first signs of water trapped in the firn. We were anticipating the water table to be near 15 meters from the airborne radar data, so there was still a way to go. But on the morning of the second day, at about 13.5 meters depth, we pulled a drill core back up and saw the barrel dripping water. Hurray!! Almost exactly where we predicted, and in complete agreement with the satellite and airborne estimates. We drilled a few more meters in the evening to confirm our finding and happily celebrated over a nice dinner in the clam tent.

We carried on drilling over the next days and to about 30 meters, we started pulling out cores made only of ice indicating that we were below the firn aquifer. We stopped drilling at about 35 meters. Following the drilling we spent a full day on hydrology measurements to characterize the firn aquifer’s physical properties, such as its permeability (using slug and pumping tests) and lateral flow using a salt dilution method.

The last step was to drop a string of temperature sensors into the borehole in order to closely monitor the thermal state of the firn for the next 3 years or so. We said good-bye to the thermistor string, and used fresh fine grain snow to backfill the borehole.

On December 12, we packed up our Wilkins camp and flew to the George VI Ice Shelf site. There, we did not anticipate much of an aquifer which was confirmed by our drilling efforts. We did a similar 35-meter hole and the borehole remained dry and cold with fewer ice lenses. Two notable half meter thick ones at about 10.5 and 30 meters showed up as reflections in the radar profile as well. The drilling was more complicated at this site since the thermal drill was mainly made to drill in warm ice and water, creating a good thermal contact. We did some repairs halfway through the drilling to replace the heat ring as the heating element had burned on the first ring. It was a short day of repair and the new heat ring performed well. We finished drilling to 35 meters and dropped another string of temperature sensors similar to Wilkins to be able to compare the two sites.

Once back at the University (both University of Colorado Boulder and University of Washington in Seattle), we will spend the coming weeks transcribing our notes, processing the firn and ice core data, and analyzing the hydrology data. After flying north from Rothera on December 20, we look forward to sharing our results with our colleagues.


This photograph shows the drilling and core processing pit. Note that the wind break was not very useful with the wind blowing in the opposite direction. Credit: Ted Scambos


The drill barrel is back at the surface after collecting an 80-centimeter firn core at about 20 meter depth. Notice the break in the ice core that is hanging at the bottom of the barrel. Credit: Ted Scambos


The next step is to get the core out of the barrel, for that we used magnets to release the “core dogs” (two little metal teeth that break the core) and the core slides out of the barrel. Credit: Ted Scambos


Lynn is cutting a core section and will be measuring its dimensions and weighing it. In the meantime, Bruce takes careful notes. Credit: Ted Scambos


For the first one and a half meters, the core sections mainly made of snow were too fragile to be measured accurately. Therefore, we dug a snow pit and collected snow-density measurements using triangle-shape snow cutters with a 10 centimeter resolution. Lynn is showing one of her a perfect cuts. Credit: Ted Scambos


It was not always a smooth ride for the drill. In this photo, we are in the process of replacing the heat ring. Credit: Ted Scambos


After we drilled the borehole, we performed various hydrological tests. For one of them, we added, at the surface, a saline solution that once mixed in the borehole was used as a proxy to quantify lateral flow in the aquifer’s water. Credit: Ted Scambos


We used a plumbing-type camera called SeeSnake to inspect the borehole. In the photo collage there are a few snapshots of what we recorded at different depths, from the surface down to about 20 meters. Credit: Ted Scambos


Happily posing with freshly collected cores! Credit: Ted Scambos

%d bloggers like this: