How a beer bottle helped reveal rapid past climate change

According to Willi Dansgaard “A sophisticated experimental set-up on the lawn became the beginning of a new field in geophysics.” Credit: Centre for Ice and Climate, Niels Bohr Institute, University of Copenhagen.

On Saturday June 21, 1952, in a garden in Copenhagen, Denmark, raindrops fell through the slim neck of a beer bottle, splattering and splashing as they hit its bottom. But the bottle wasn’t carelessly left behind – Willi Dansgaard had inserted a funnel into its neck so he could use it for an experiment. He was watching it closely, collecting rain to later measure in his lab. Each drop brought Willi closer to revealing the secrets of Earth’s history, by giving scientists a way to work out temperature from ancient ice. In doing so, he would help show how climate can change much faster than experts had thought possible.

Willi was born in Copenhagen in 1922, living and studying physics and biology there until going to work for the Danish Meteorological Institute (DMI) in 1947. The DMI sent Willi and his wife Inge to Greenland, first to study the Earth’s magnetic fields, and then to help improve the reliability of weather forecasts. Their time there left the pair with ‘deep impressions of the course of Greenland nature, its forces, its bounty, its cruelty, and above all its beauty,’ Willi wrote in his autobiography. ‘We were both bitten with Greenland for life, but after a year the need for further education forced us to turn homeward.’

So in 1951, Willi took a job at the biophysics research lab at the University of Copenhagen, where his first job was to install a mass spectrometer. Able to distinguish between chemicals using weight differences, mass spectrometers are often described as atomic-level weighing scales. But they actually measure molecules’ weight by firing them through an electromagnetic field at a detector, similarly to how bulky old TVs fire electrons at their screens. Though mass spectrometers existed since the early 20^th century, Second World War US efforts to produce uranium for an atomic bomb had boosted their power. Willi set up the type of machine that had been invented in the course of that work, so his department could detect tracers used in medicine and biology.

By 1952, Willi knew that his mass spectrometer could separate forms of the same chemical elements – or isotopes – that could differ in weight by as little as a single neutron. And faced with a wet weekend in June, he wondered whether the amount of these isotopes in rainwater could change from one shower to the next. ‘Now when I had an instrument that ought to be able to measure it, there was no harm in trying,’ he writes. ‘I placed an empty beer bottle with a funnel on the lawn and let it rain.’

One good idea

Changes in the amount of oxygen-18 (δ) in vapour evaporating from the sea and rain and snow that forms as a cooling air mass moves towards higher latitudes and/or higher altitudes. As more oxygen-18 is rained out than normal oxygen, the cooler it gets the less oxygen-18 is left in rain and snow. Credit: Centre for Ice and Climate, Niels Bohr Institute, University of Copenhagen.

The Copenhagen rain came from an unusually well-formed boundary between a warm front and a cold one, and when Willi analysed his samples he saw clear changes. The amount of a heavier form of oxygen – oxygen-18 – in the rainwater dropped sharply as the boundary passed over him. Rain was forming at a higher, colder level in the atmosphere and the temperature was affecting the amount of oxygen-18 in the water that fell.

The colder the cloud, the less oxygen-18 there was in the rainwater, Willi found. He proposed that this was because the heavier water molecules are less likely to evaporate and more likely to condense into rain. Water vapour mainly evaporates to form clouds near the equator that move towards the poles. The rain falling from these clouds has more water containing oxygen-18 than the vapour left behind. As cloud progresses polewards more rain falls, leaving ever smaller proportions of oxygen-18 in the water vapour. Willi’s explanation for his beer bottle measurements relied on the idea that cold temperatures speed up this process.

Willi then set about working out the details of this cycle, a task that would take over a decade to fully complete. At first he used samples from his friends in Greenland and river and tap water sent from all over the world. The biggest boost came when the International Atomic Energy Agency and World Meterological Organization started collecting data on oxygen 18 levels in rainwater collected in the early 1960s. Data from 100 stations showed a similar pattern to what Willi had first seen in his garden – rain and snow in colder climates contained lower levels of oxygen-18.

This discovery opened a scientific opportunity, Willi realised early on. If he could recover records of old rainfall, Willi wrote, they ‘might reflect the climate at the time of formation of the water. Now, where do you find old water? In glacier ice. And where do you find old glacier ice? In Greenland. This is how my interest in Greenland was revived, now in a new context. I was sure it was a good idea, maybe the only really good one I ever got.’

And so alongside establishing the reliability of the oxygen-18 method, Willi started collecting samples to test this idea. In 1958, he sailed from Norway on a fishing boat, as part of a team that collected and melted icebergs from Greenland. At the same time, European scientists were on a mission to drive tracked vehicles across Greenland and drill cylinders of ice, or ‘cores’ 10-20m beneath the surface. Willi asked them if he could analyse ice samples they collected. When he did he was able to show that, like the rest of the world, Greenland had warmed from 1920 to 1945.

Quick change

‘I served as a coolie and a packing master,’ writes Willi Dansgaard (shown smoking his pipe here) of his Dye-3 ice core drilling mission to Greenland in 1973. Credit: Centre for Ice and Climate, Niels Bohr Institute, University of Copenhagen.

Also in 1958, the US was carving a military research station out of the Greenland ice, called Camp Century. After drilling a number of shallow ice cores in the early 1960s, scientists there developed a thermal coring drill able to collect cores all the way down to the bedrock underneath the ice sheet. They drilled a 1,387 metre core between 1963 and 1966 that contained more than 100,000 years of climate history, back to the start of the last ‘ice age’.

During a research expedition to Greenland in 1964 Willi was able to visit Camp Century, but he wasn’t allowed to see the drilling. Instead he had to make a proposal to the scientist responsible for the ice cores, Chester Langway, to analyse the whole length for free. That offer became a historic event for climate science, as when Chester agreed, he, Willi, and their teammates produced the first long-term temperature record from Greenland.

Their analysis showed a large temperature change around 11,000 years ago at the end of the last ice age and sudden, big, temperature swings before that. That shocked scientists at the time, who generally thought climate changes happened gradually over long time periods. In later years, Willi would return to Greenland to extract more ice cores together with Hans Oeschger from the University of Bern, confirming these findings. The abrupt climate changes during the last ice age are now known as Dansgaard–Oeschger events.

Though today Willi’s oxygen analysis is teamed with other methods, by the time of his death in 2011 he had played a leading role in setting up ice-core science. Building on his work, scientists continue to study these natural popsicles to improve our understanding of how and why climate has changed through history. And his discovery of rapid temperature changes helped awake scientists from their complacency about global warming. The legacy of Willi’s simple back-garden experiment lives on in the knowledge that climate can change dramatically – and that we could make that happen.

The profile of changes in the oxygen-18 level (δ) in the Camp Century ice core, plotted on a depth scale (left) and a preliminary age scale (right). The black and grey areas correspond to periods when Greenland temperatures were higher and lower than the time of the study, respectively. The large grey area reflects the last glaciation, or ‘ice age’. The heavy line in the lower right corner marks an age of 100,000 years according to a more advanced ice flow model. The lumpiness of the glacial period at the bottom shows rapid temperature changes that were a shock to scientists when Willi’s team revealed them. Credit: Centre for Ice and Climate, Niels Bohr Institute, University of Copenhagen.

Further reading:

Spencer Weart’s book, ‘The Discovery of Global Warming’ has been the starting point for this series of blog posts on scientists who played leading roles in climate science.

Willi Dansgaard’s autobiography Frozen Annals is full of many more entertaining anecdotes than I could include here.

Willi Dansgaard (1953). The Abundance of O18 in Atmospheric Water and Water Vapour Tellus DOI: 10.1111/j.2153-3490.1953.tb01076.x
Willi Dansgaard (1954). The O18-abundance in fresh water Geochimica et Cosmochimica Acta DOI: http://dx..org/10.1016/0016-7037(54)90003-4
Willi Dansgaard (1964). Stable isotopes in precipitation Tellus DOI: 10.1111/j.2153-3490.1964.tb00181.x
W. Dansgaard, S. J. Johnsen, J. Møller, C. C. Langway Jr. (1969). One Thousand Centuries of Climatic Record from Camp Century on the Greenland Ice Sheet Science DOI: 10.1126/science.166.3903.377