by Sarah Dry
The answer to the implicit question raised by Milankovitch’s work—What traces would such cycles have left on the earth?—led to a new field. This field, called paleoclimatology, combined the descriptive elements of climatology with new physical tools to generate a map (or series of maps) of the earth’s past climates. The location and nature of rocks had long served as crude proxies for temperature. Using the geological traces—the literal scratches left by glaciers as the ice passed over bedrock, or the moraines deposited when they melted—it was possible to infer that temperatures had been low enough to support ice or warm enough to melt ice. Any finer degree of resolution was impossible. To see with more acuity, it would henceforth be necessary to make a literal rather than figurative journey into the earth’s past, to drill quite literally down into the planet to extract from ice sheets the cold remnants and from muddy ocean sediments the undisturbed detritus of deep time. These cold and muddy archives could only be read in the 1950s, thanks to new tools that relied on the same physical understanding that had built the atomic bomb. Paleoclimatology was built largely (though not entirely) on the foundations of nuclear physics and came to fruition in the postwar world of anxiety and optimism engendered by the power of the bomb.
* * *
Having come up with his idea for an isotopic thermometer for the past, Dansgaard was gripped by the simultaneous urge to share his idea and to protect it. He wanted priority, but in order to stake his claim, he also had to publish it, or at least enough of it to make clear what it was. The problem was that he hadn’t yet proved it would work. So he wrote up his paper in which he had suggested vaguely that it might be possible to use his technique to peer into the past “several hundred years” by analyzing Greenland glaciers.9 He mentioned Greenland because of the wondrous fact that the snow there never melted in the summertime. A record of each year’s snow was preserved under the next year’s snowfall, creating an ice sandwich several kilometers thick in which every layer represented a distinct year. Though he did not know it at the time, this record stretched back not hundreds, or even thousands, of years, but tens of thousands of years.
Dansgaard set off to find some old ice. He figured his best bet would be to catch icebergs as they calved from glaciers into the sea. He started in Norway, with a charismatic Norwegian named Pete (Per) Scholander who believed that air bubbles trapped in the ice were tiny atmospheric time machines, holding samples of the air from the time when the snow fell around them. Not only could the bubbles be made to speak of their contents—what mix of atmospheric elements such as nitrogen and oxygen was present in the past—but, thanks to the new technique of dating the CO2 content in the bubbles using the carbon-14 isotope, they could tell the age of the atmosphere they contained. It was a perfect fit for Dansgaard’s interest in using oxygen isotopes to study past climates. Together, Dansgaard and Scholander set out for the Jotunheimen massif. Within six weeks they’d managed to pry two very large pieces, each weighing five tons, from the old and young ends of the glacier. They melted the ice and captured the released air. Carbon-14 dating gave its age as 700 years old, in agreement with estimates by glaciologists. But there was an unforeseen problem with the relatively young snow of the Norwegian glacier. As the water melted, it took with it some of the most soluble gases, skewing the ratios of the gases in the atmospheric air. They needed colder ice, untouched by meltwater, to avoid this problem.
And so one year later, Dansgaard finally found himself heading back to Greenland, on a much more involved expedition with Scholander, which he dubbed the Bubble Expedition of 1958. They sailed around Cape Farewell at the southernmost tip of Greenland to the western coast. There they set anchor and erected a bulky laboratory on the front of their boat (to the dismay of the captain). They waited for the giant glaciers on the coast to calve into the sea and then set about spearing chunks as big as they dared, harpooning the great white blocks as if they were whales. Or they rammed their ship into smaller icebergs and hoped for pieces of the right size to result—holding on to fragile glassware in the shipboard laboratory as they did so. It was a scientific adventure, full of improvisation and high spirits.
They melted ice day and night, and their “bubble” harvest from a summer spent herding icebergs was eleven samples of carbon dioxide coaxed from the bubbles in the heart of the glacial ice, each a distillation of some six to fifteen tons of ice. The science, and the effort, seemed designed to reduce maxima of ice and effort into minima of carbon dioxide and data. This suited the men, who were seeking an elusive signal from the past. Clearing away masses of extraneous material was just what was needed to get to the signal they wanted to capture.
Dansgaard himself returned with not dozens but thousands of plastic bottles, each filled with a melted ice sample, a library, or frozen annal as he called it, of the earth’s past climate. He put his mass spectrometer to work on them and soon had a mini-production line going, producing twenty isotopic ratios, or delta values, a day.
Ice looks different when it has been flooded with meltwater, and Dansgaard could tell just by looking at the sample of the glacier ice in an ice chunk they studied which layers had meltwater and which didn’t. His delta values matched up to these visible layers. It was another good sign for his good idea, but the ice they’d sampled wasn’t old enough to prove what Dansgaard really wanted to know: Could temperatures be read off the oldest ice in Greenland? Answering the question would require getting off the water and into the interior of Greenland, where the ice sheet was deepest. Hundreds of meters beneath the surface of the ice lay the records of the past climate. Dansgaard was convinced of it.
* * *
As Dansgaard raced about trying to collect samples from rivers, storm clouds, icebergs, and glaciers around the world, his mass spectrometer machine sat in Copenhagen, the ultimate destination of the water that Dansgaard collected and the reason it was even worth doing so. The machine did not, in fact, analyze the water samples themselves. It was easier to transfer the oxygen molecules from the water into a bit of gas, carbon dioxide, and then analyze that by causing the molecules to diverge according to weight.
The water that Dansgaard collected contained more than just a ratio of isotopic variants. When he stood out in his back garden collecting rainwater in 1952, he was also collecting traces of radioactive elements released into the environment—the atmosphere, the ocean, and the earth itself—by the explosion of atomic weapons. Aside from the two bombs dropped on Japan by the United States, these weapons had been exploded not to wage war—at least not directly—but to prepare for the possibility of war by testing the effects of different types of weapons.
Both the bombs and the mass spectrometer Dansgaard used to trace the isotopes were evidence of the success of a new kind of physics, the physics of the nucleus of the atom, which had grown out of the discoveries made some fifty years earlier that the atom was not immutable but mutable. Atomic physics had shown that the atoms too, had histories, and sent out trajectories of change over time in the form of packets of energy called radiation. Change, it seemed, was built into the very fabric of matter.
Some forms of change were, however, more natural than others. Radioactive isotopes had been entering the earth’s environment in ever-greater numbers, since the Trinity test and the detonation of the bombs over Nagasaki and Hiroshima in 1945. These were not the stable, “natural” isotopes that had been present in the earth from its origins. These were the by-products of atomic explosions, unstable radioactive isotopes with unfamiliar names such as strontium-90, plutonium, iodine-129, caesium-125, and tritium. They threw off dangerous packets of excess energy that destroyed the fragile cells of living creatures. Acute radiation sickness could follow immediately, or the effects could be delayed, manifesting years later as raised rates of cancer and genetic abnormalities in the offspring of those exposed.
Atomic testing increased throughout the 1950s as tensions between the United States and the Soviet Union rose. By the e
nd of the decade, much damage had been done by radioactivity to living creatures in the environment. Exactly how much, no one knew. To answer this question, the International Atomic Energy Agency (IAEA) partnered with the World Meteorological Organization (WMO) to try to find out exactly how much man-made radioactivity was circulating, and where. In order to do so, they established just what Dansgaard needed—a global network for rainfall collection. Such global projects had roots extending back to 1853, when Matthew Fontaine Maury, the head of the U.S. Naval Observatory, called for an international system for monitoring the weather over the land and sea. In 1905, Léon Teisserenc de Bort proposed that the International Meteorological Organization (the precursor to the WMO) create a Réseau Mondial, or global network, for meteorological observations. Both had faltered under the weight of their own ambitions. Small differences in national observing practices added up to major inconsistencies in the data. Some countries, for example, took meteorological measurements at three-hourly intervals, while others insisted on two-hourly intervals. Differences in instrumentation, and varying guidelines for locating and reading instruments, compounded the problems. Even if data could be meaningfully compared, the sheer scale of the project was overwhelming and much data was left raw and unreduced.10 More successful were a series of dedicated observing “years” which focused the efforts of scientists for a limited period of time. Two International Polar Years (held in 1882–1883 and 1932–1933) had laid the foundation for a more ambitious International Geophysical Year (IGY) in 1957–1958, which galvanized more than sixty-five nations, including both the United States and the Soviet Union, in a rare moment of scientific détente, into an eighteen-month frenzy of studying the earth from every imaginable angle.
In comparison with the IGY, the IAEA-WMO collaboration was narrowly focused, aiming solely to track the spread of tritium, a single radioactive element released by atomic testing. Soon, more than a hundred stations around the world were equipped to collect precipitation samples and send them to a central office in Vienna. The same water samples that they collected also contained—as did any water sample on earth—a distinctive ratio of oxygen isotopes. Dansgaard got wind of the project and seized another golden opportunity to piggy-back on the global reach and deep pockets of an international organization. All he needed was a few milliliters of the water to do his analysis. He mobilized a Danish contact on the IAEA and thereby hitched a ride on this great global rain hunt. It was not the first or only time that military matters would influence the direction of new science. As ever, Dansgaard proved adept at taking what he needed from such well-funded enterprises.11
The network spread itself as evenly as geography and politics would allow. South and North America were covered from Barrow, Alaska, in the north to the Falkland Islands in the south. Africa was speckled with stations, as were Europe and Australia. The great conjoined expanse of the Soviet Union and China remained an unrepentant blank. That left a large hole in the global precipitation network, but there were enough samples from elsewhere that a good understanding could be gained. Dansgaard was soon swimming in samples—a hundred per month (one from every station) at one point. The mass spectrometer machines—a fancy French one for measuring deuterium and the old standby from 1951 for the oxygen isotope—were run day and night.
To be useful, a thermometer must be consistent and give the same reading for the same temperature no matter where it is. Water evaporates and condenses in complex ways as it travels around the globe. It wasn’t clear that the processes that made cold air isotopically heavier in Copenhagen would hold true at every location. Sampled water had a “pre-history” of multiple condensations and evaporations which determined how much of a given oxygen isotope it contained. Would a consistent relationship between the percentage of heavy water and local temperature hold for water samples collected around the globe, each with vastly different prehistories? The answer, with some caveats, was yes.
The IAEA-WMO data showed that it was possible to use the oxygen isotope technique to trace the “circulation patterns and mechanisms of the global and local movements of water,” as Dansgaard put it in the paper he published on the results.12 More than a decade earlier, Dansgaard had begun collecting rainwater in his back garden in Copenhagen. He’d shown that he could peer into the heart of a single storm using his mass spectrometer. Now he had shown that oxygen isotopes bound in water could reveal global patterns of evaporation and condensation, the flow of water molecules that moved heat around the earth and drove both ocean and atmospheric circulation. As temperatures fell at the stations whose ratio of oxygen isotopes he could measure, he found that there were more and more of the isotopically heavier atoms. The old correlation—between heavier isotopic weights and colder temperature—held in a variety of locations across the globe. He included a graph showing how the samples from a range of Pacific islands lined up, with the neat correlation between temperature and 18O ratios holding across thousands of miles.
Dansgaard had now shown that the ratios of oxygen isotopes could be used as a thermometer, but he still wanted to know if they could be made into a time machine. The very radioactive fallout that had helped his progress, in that it had spurred the IAEA-WMO cooperation, now intervened. As a follow-up to his second Bubble Expedition, he was working on the problem of using a radioactive isotope of silicon, known as Si-32, to date ice, but the samples he’d taken from icebergs were all contaminated with fallout from Soviet tests. Dansgaard now turned to a peculiar and fascinating location—and project—that was literally underway on Greenland. Located on a portion of the island that lay directly on the flight path between the eastern United States with western Russia, Camp Century was an experimental U.S. Army station dug under the snow. Located inland just over one hundred miles (thus the name) from Thule Air Base, Camp Century became a unique piece in the puzzle of Cold War geopolitical strategy.
For the U.S. military, understanding the Arctic environment was critical to operating successfully within it.13 One army official noted that although it was unusual for military researchers to have to “reach so far down toward the foundations of science as it has to do in this exceptional case,” knowledge of snow and ice was so minimal that much basic research was necessary.14 Camp Century was run by the Snow, Ice, and Permafrost Research Establishment (SIPRE), a Defense Department laboratory set up in 1948 to prepare the U.S. military should it need to mount an offensive across Greenland. Could heavy airplanes land on snow-compacted runways, or even floating sea ice, in order to supply stations along a planned network of fifty Arctic radar stations? Would it be possible to launch nuclear weapons from under the ice? Could a ground campaign be launched across the frozen expanse? Was it feasible to build and run a railway under the snow, to transport goods and men?15
By 1964, the U.S. Army had been actively trying to answer these fantastical questions for five years. They had built a futuristic sub-snow encampment on a scale never attempted before or since. Two hundred men lived in a series of tunnels below the snow, complete with a 4,000-book library, laundry, mess hall, barber shop, and hospital, and liberally supplied with hearty cuisine including steaks, green beans, and mashed potatoes. It was all powered by a nuclear reactor located only 300 feet from the men’s living quarters.
Dansgaard was largely unimpressed by Camp Century’s operational ambitions (he noted the folly of attempting to build a railway under the snow—it had started to buckle and bend under the pressure of the ice as soon as it was begun), but it could provide him his much-needed untainted snow, preferably snow which was just older than the tests which had released Si-32 into the atmosphere. The depth to which Camp Century was dug had exposed just this sort of snow, so he set out in the summer of 1964 to collect some samples. He duly gathered the ice he needed, noting the weird life under the ice, with its mix of American cheeriness (drinks cost just twenty-five cents regardless of size) and brutal extremes of cold and pressure.
FIG. 7.4. The thermo-drill set up at Camp Cent
ury, 1964. Dansgaard never saw the rig during his visit.
The few days he’d spent at Camp Century he’d been busy with his own work. Just meters away, on the other side of a wall of snow and ice, an enormous drill was coring deep down into the thick cap of ice which covers Greenland. Dansgaard left Camp Century without ever seeing the rig, which was a military secret. Over the course of half a dozen years, the Americans drilled deeper into the ice than anyone had ever managed to go. The practical challenges of drilling so deep were substantial, with the walls of any drill-hole subject to enormous pressures. The tool that did the job was a special thermo-drill that simultaneously melted its way through the ice and preserved a core of frozen ice uncontaminated by meltwater. It was so expensive that only the U.S. military could afford it. By 1966, the drillers had hit bedrock, 1,390 meters beneath the surface of the ice.
Once Dansgaard got wind of the long core, he knew it would be just as valuable for studying Arctic climate as the IAEA-WMO samples had been for studying the water cycle. Dansgaard had proven his ability to use isotopes to study samples of both ice and liquid water to study climate in the present and the past, and it wasn’t long before he’d persuaded the right people to give him some samples from it. Though neither he nor the U.S. military knew it yet, by far the most enduring and valuable product of six years and countless dollars would be the result of what Dansgaard and his colleagues did with these samples. What he did would generate data that showed just how varied the earth’s climate had been in the past, and, just as importantly, how abruptly the climate could change.
