"We were shocked how far advanced the melting is already"

Ms. Schwikowski, you wanted to drill 80 metres deep into the glacier, right down to the bedrock, to recover even the oldest layers of ice. How did this plan turn out?

Unfortunately, it was a wash. Already in 2016, our colleagues from Venice had extracted a ten-metre-long ice core from the Grand Combin. It looked promising and showed the expected signature from annual fluctuations in particle deposition. And so, together, we actually wanted to extract a longer ice core last year to save it from the melting of the glacier, for future generations of researchers. Bad weather and the coronavirus pandemic saw to it that we could only make the expedition now. And we had barely started drilling when we ran into trouble.

What was the problem?

After only half a metre we encountered a hard layer of ice, where we normally expect relatively soft firn – firn is old snow that has been transformed into sleet-like grains. In alpine glaciers it normally only turns into a compact ice under pressure at a depth of 40 to 50 metres. Above that, we call such a layer superimposed ice. That occurs when it's so warm that the snow on the surface melts and the water seeps in to a depth where it is cold enough to freeze again. So that was not a good sign.

But was it possible to continue drilling?

Yes, although we kept running into such layers. Then at 17 metres the drill got stuck. It has a hollow aluminium core barrel almost a metre long. This is lowered, along with the chips barrel and the motor, on a cable that supplies it with power. We use it to cut an ice core eight centimetres thick and 70 centimetres long out of the glacier. Then we take out the drill, store the ice core it contains, and drill the next piece. And now, while pulling it out, we got stuck. Probably because the solid layer of ice at 17 metres had shifted, or because the layer underneath was so soft that the drill could easily move sideways and got jammed against the ice when it was pulled back up. With a lot of effort we were able to fish it out, and we realised that there was even liquid water down there. When we tried again, we finally got stuck at 25 metres. We were only able to free the drill again with antifreeze. This of course made that borehole unusable – the chemicals would have contaminated every further sample.

Did you try to drill in other places?

Naturally, several. But everywhere we got stuck again at the same depth. So the compacted areas were not just spatially limited ice lenses, but entire layers running through the glacier. We had to stop, because otherwise we would have risked losing the drill.

How did the team interpret that?

We were totally shocked. Because that made it clear: For this glacier, we may already be too late with our project. Here it must have been so warm in recent years that a great deal of meltwater was able to penetrate far into the depths. When that happens, it transports the very particles whose stratification over time we want to retrace. If it is too warm for a few days and stays cold the rest of the time, that's no problem. Then the meltwater does not reach a great depth before it freezes again. Maybe only the topmost annual layer will be affected. This is something we experienced in Tibet and Siberia. But the situation in the Alps looks bleak. We can handle a few gaps, but not a signature that's distorted throughout.

What happens now?

We do, after all, have some 17-metre-long ice cores that cover the last 15 years or so. That's not much, but we hope it may be enough for one particular research project: One of our doctoral candidates would like to examine the extent to which organic molecules, for example from forest fires or from the burning of fossil fuels, have been preserved in glacier ice. So we are now analysing whether at least these 17 metres are usable. If so, then maybe we can try again with a thermal rather than an electromechanical drill. This type of drill melts its way down and does not get stuck so easily. We didn't have it with us because the reconnaissance study  didn't give us any reason to expect problems. However, if the ice core proves to be unusable, the only alternative in the Alps is the Grenzgletscher in the Monte Rosa massif, southeast of Zermatt. The Colle Gnifetti saddle there is at 4,500 metres and should not have been affected by the melting – at least that was the case when we last drilled there in 2015.

Can you speed up the worldwide extraction of ice cores in time to save Ice Memory?

Our international initiative aims to collect a good 20 ice cores from all over the world. With the Grand Combin we would have five of them – including Mont Blanc, Elbrus, Illimani in Bolivia, and Belukha in Siberia. There is also a plan to drill on Kilimanjaro in East Africa, but the political situation there is currently too difficult to get started. And there lies one of our obstacles: We can't simply go faster. On the one hand, because the circumstances on many glaciers are difficult and we have to wait for the right time. On the other hand, because these expeditions are time-consuming and can only be carried out by experienced teams. The 20 or so researchers involved do this more or less in their free time. We always have to combine this with other research projects so that we can justify the effort. That is why we now want to set up an international foundation, of which PSI is a founding member, to put the project on a more stable financial footing. The global ice core community wants to meet at a conference in Crans-Montana next year. There one topic to discuss is the progress of Ice Memory.

Interview: Jan Berndorff