How an Ancient Ice Age Froze the Entire Earth—And Helped Humanity Flourish

When you think of ice, maybe your mind first goes to an object. An ice cube, an icicle, a slushy. Something inert, a thing that can be handled. At a certain scale, ice does behave this way. But pile up enough of it, enough to create a glacier, and this material takes on its own agency. What once sat blank and unmoving begins to flow and creak and jerk beneath its own weight. Stress and tension crack into crevasses that yawn deep enough to end lives. These forces also spill out beauty, the most brilliant blues that glow like sapphires or hard candies. 

We live in a time of pity for glacial ice. It’s falling apart, it’s melting, it’s destabilizing. It’s retreating, giving in to warmth like prey to a predator.

These words hold truth, but they also obscure the strength of this substance. Glacial ice is not simply a passive product of the planet’s climate; it is an engineer of it, and a ruthless agent of change. 

We owe our own existence, in certain ways, to a series of ancient ice ages that spanned between 717 million and 635 million years ago. Strata from this geologic period tell us that life as we know it today may have bloomed in the wake of Earth’s coldest, hardest times. 

*

Remnants of these ancient ice ages rest on the mountainous slopes of Svalbard, a Norwegian archipelago located roughly 1,000 kilometers from the North Pole. A young British geologist named Brian Harland discovered these strata while out mapping that icy terrain last century. A photo captured during the first of more than forty expeditions he would take shows a scruffy-haired twenty-one-year-old with a wide, toothy grin and high cheekbones that frame earnest eyes. With shoulders pushed back and head leaned toward the camera, he looks totally in his element amid those mountains and fjords.

We owe our own existence, in certain ways, to a series of ancient ice ages that spanned between 717 million and 635 million years ago.

That was 1938. The outbreak of World War II the following year would prevent him from returning for some time. When he finally did make his way back a decade later, he did so as a professor of geology at the University of Cambridge, bringing a cadre of students along with him.

As Harland and his students made their way across Svalbard’s frozen landscape, they pulled together evidence of another icy time that had taken place on that land hundreds of millions of years earlier. They found a few telltale signs of glacial activity in the ancient strata that they recognized as such thanks to their understanding of glacial activity in the modern world. For example, whereas rivers of water struggle to move debris larger than dinner plates in diameter, rivers of ice pluck up stones the size of golf carts and push them along like dust in a dustpan. These boulders get mixed in with much finer sand and clay to form what’s called a till. The randomness of glacial till distinguishes these deposits from those laid down by more delicate and organized movements, such as the flow of water through streams or out in the open ocean. Harland and his students recognized glacial till because it lay in plain sight among modern glaciers. 

Unsorted mixes of sediments can also accumulate from other sorts of geologic activity, like landslides, and so they looked for a suite of other clues in the strata to weed out alternatives and affirm they had, indeed, found evidence of ancient ice. 

When glaciers move across the ground, they drag rubble beneath them, carving shallow, parallel grooves called striations through the bedrock. When icebergs drift out to sea, they melt out rubble that drops to the seafloor as pieces of debris called glacial dropstones. Because dropstones fall from above, they can pierce and depress the fine silty layers of the seafloor and cause those layers to warp around them, like a sofa cushion hugging a body.

Of all the visual clues used to identify glacial deposits in the rock record, dropstones often provide the most convincing evidence of ice. Few other forces can send rocks dropping from above in a body of water. Uprooted trees can carry stones in their roots and release them as they get caught up in a current, but the strata Harland was studying had formed hundreds of millions of years before trees had even evolved.

Check, check, and check. All of these signs of glaciers present in the modern sediments of Svalbard also appeared in the ancient strata. Harland understood that ice ages come and go, so evidence of ancient ice probably didn’t surprise him very much. But what did surprise him was what he and his students found interspersed among the glacial deposits: thick layers of yellow and gray carbonate rock. 

Today, carbonates form the foundations of Australia’s Great Barrier Reef, the Bahama Banks, and other such tropical locations. They typically form only in the world’s warmest waters, which rush around the equator. If their interpretations were correct, Harland and his students had found evidence not only that Svalbard had once been closer to the equator but that, at some point while it was there, it had also been encased in ice.

This, Harland realized, was odd. In today’s world, glacial ice can persist in equatorial regions such as the Andes and Indonesia, but only on mountainous slopes several miles above sea level. If ice had formed all the way at sea level in the warmest part of the world—the part with the most direct sunlight, around the equator—then ice probably would have smothered much of the rest of the planet as well. If that were the case, this would mark the most extreme ice age known in all of Earth’s history. At no other time had glacial ice made it to sea level that far from the poles. The most recent ice age that peaked some 20,000 years ago covered only about 8 percent of the planet; Harland’s ice age looked like it covered closer to 100 percent (though he wouldn’t come to this conclusion until years after his initial observations).

From outer space, this cryogenic version of Earth would have glistened white and bright. Beneath the surface of the frozen ocean, dark and briny currents would have stirred continuously from geothermal heat billowing out of cracks in the seafloor. Up above, winds would have howled around the equator in strong gusts, and tiny ice crystals would have settled out of air that remained below freezing everywhere. Life persisted through all this, but it couldn’t have been easy. Algae, bacteria, and simple protozoa would have squirmed around on top of or under the ice, and other specks of life would have swarmed alongside hydrothermal vents on the seafloor.

This all may seem a bit fantastical, but Harland couldn’t see any other way to interpret the Svalbard strata. When he and colleagues conducted follow-up studies of magnetic minerals within these deposits, they found that those grains—which behave like fossilized compass needles—pointed to a near equatorial origin for the rocks as well.

“It is concluded,” he wrote in a 1964 paper documenting his unusual finds, “that an ice age was sufficiently extreme to form marine tillites in the tropics.”

That is, glacial tills. In the oceans. In the tropics. 

Energized by these findings, Harland presented his theory to colleagues. But it was a hard sell. Rather than accept his seemingly fantastical ideas, some wove together alternative theories. They suggested that those sediments came not from ice but instead from something more mundane, like a series of widespread and robust mudflows. But Harland persisted. Mudflows, he argued, don’t drop uniformly over thousands of square miles, and they don’t leave behind dropstones. 

Further to his point, he wasn’t the only geologist finding evidence of ancient glaciers in rocks of roughly this age. As early as 1891, a Norwegian geologist reported deposits of a similar age along the shores of a fjord in northeastern Norway that had striations—those long grooves that form when glacial ice sandpapers the ground beneath it. Over the intervening decades, Harland and others identified some two dozen localities of similar randomly sorted, striated, dropstone-covered rocks across the world in deposits of a similar age, from East Asia to Western Africa to South Australia and Greenland—many with the same curious layers of carbonate capping them off. 

Individually, the glacial interpretations of these rocks didn’t raise many concerns. But when they were taken all together, and the warm-water carbonates were thrown in, the story became harder to stomach.

These glacial tills weren’t simply a blip here and a blop there. They were substantial. They measured between ten to one hundred meters thick, sometimes with boulders as large as baby rhinos. To Harland’s mind, these were all too expansive and widespread to have come from mudflows. The suggestion that they had was more outlandish to him than the possibility of ice at the tropics.

Getting to the bottom of this global cooling event might reveal how the planet produced the complex multicellular beings that would one day evolve into us.

The mudflow rebuttals drifted away over time, but many of his colleagues still couldn’t accept the idea of ice covering the turquoise waters and soft sand beaches that they thought of as the tropics. They couldn’t imagine a version of Earth that could ever get that cold, nor could they see how the planet would spin out of such a deep chill. They couldn’t, it seemed, grapple with the possibility that Earth was once much different than it is today. 

These colleagues continued to come up with more alternatives. Some called for wandering magnetic poles that would have shuffled across the planet and brought cold snaps directly to continents, like a waiter serving drinks around a dining room. But that hypothesis, Harland wrote in a bristly 1964 Scientific American article, was not supported by any concrete evidence. “Indeed,” he and his coauthor, Martin Rudwick, lamented, “it seems to have been suggested only in order to avoid postulating the presence of ice in the tropics.” 

Harland and Rudwick were not willing to be so avoidant. “Failing any other explanation,” they wrote, “we are prepared to accept that [ice] did exist there.”

That those brilliant turquoise waves had flattened and hardened into muted whites and grays. That smells of salt gave way to ice’s absent scent. 

The desperation with which Harland worked to understand this ice age stemmed not only from its strangeness but, perhaps even more so, from its timing in relation to another major event in Earth history. Relatively soon after the last of the ice deposits drift away, a burst of fossils appear around the world. This marked the beginning of the Cambrian period, when the majority of animal groups alive today first appear in the rock record. 

“It can hardly be mere coincidence,” Harland and Rudwick wrote in their 1964 article, “that a geological event of such intensity was followed, after a relatively short interval, by a biological event of equally striking character.” 

To their minds, the ice age could have instigated the explosion of life that followed. Getting to the bottom of this global cooling event might reveal how the planet produced the complex multicellular beings that would one day evolve into us.

__________________________________

Adapted from Strata: Stories from Deep Time by Laura Poppick. Copyright © 2025 by Laura Poppick. Used with permission of the publisher, W. W. Norton & Company, Inc. All rights reserved.