Devastating Ice Age Floods That Occurred in the Pacific Northwest Fascinate Scientists

The Scablands were formed by tremendous and rapid change, and may have something to teach us about geological processes on Mars

Palouse Falls
A view of a Palouse Falls in Palouse Falls State Park in Washington. Geologists believe massive floods carved out this canyon and others in the Scablands. Glenn Traver via Getty Images

The Earth seems to change slowly. Continents shift by about half an inch in a year. Sea levels rise by less than a quarter of an inch in the same amount of time. Mountains are constantly being eroded but, to us, seem to stand today just as they did yesterday and the day before. Our planet’s geological history often seems like one of slow, grinding change. But that’s hardly the whole story. Sometimes geological change comes startlingly, violently fast, leaving scars on the Earth’s surface. The Channeled Scablands of the Pacific Northwest, a landscape full of flat-topped plateaus that rise between steep-walled canyons, are among the vastly-altered landscapes that have caused researchers to rethink what they previously presumed. The geologic wounds are dramatic evidence that quick and catastrophic changes have played a significant role in shaping our planet.

The Scablands, principally located in southeastern Washington state, bear the signs of an incredible Ice Age event. Between 14,000 and 18,200 years ago, huge glacial lakes on the boundaries of ice sheets burst from their natural dams and rushed over the landscape, scouring the hills and dropping massive stones as they went. Entire hills were washed away as the floodwaters dumped gravel, boulders and sediment in new places, almost like shaking a great geological Etch-a-Sketch. But this is a relatively new understanding, only broadly accepted since the 1970s. It took decades for geologists to construct even an outline of what the Scablands represent, a realization that proved to be a turning point for science. For if intense floods could carve such features once in Earth’s history, surely they could have changed landscapes at other times and in other places—even those as far away as the Martian surface.

Geologists only began to pick up on the story of the Scablands a century ago. In the 1920s, naturalist J Harlen Bretz wrote several descriptive papers on the strange basins and odd channels of the area. Those channels had been created by moving water, but the way water had once flowed through the area seemed to make no sense. “The channels run uphill and downhill, they unite and they divide, they head on the back-slopes and cut through the summit,” Bretz wrote, “they could not be more erratically and impossibly designed.” The only reasonable conclusion, Bretz proposed, was that the Scablands were created by massive and short-lived floods.

Bretz’s colleagues were not ready for such a conclusion. Ever since geology came into its own as a science in the 19th century, much of the field has been influenced by the concept of uniformitarianism—that the present is the key to the past. In a broad sense, that’s an excellent rule. The Earth is still changing, and many of those alterations—from erosion to volcanic eruptions—also occurred in the past. But some additional stipulations to the older formulation of the idea were unnecessarily taken as truth. One of them was that the Earth is changing at a slow, gradual rate and that quick, catastrophic change was impossible. Bretz’s idea for how the Scablands formed flew in the face of what many geologists accepted. Channels were carved over long periods of time by rivers, other geologists thought, not by sudden floods.

To Bretz, the evidence was unmistakable. Among other things, the Scablands contained layers of gravel hundreds of feet high. Slow-moving streams couldn’t have left such vast accumulations. Pieces of gravel are larger and heavier than particles of sand or silt, requiring faster-moving water to pick up the gravel and transport it. Gravel deposits as tall as skyscrapers must have required an incredible amount of fast-flowing water. The pattern was consistent with the underlying geology of the area, as well. The rock beneath the massive flood deposits was relatively friable volcanic rock, easily broken and carved. The fragility of these rock layers allowed the floods to gouge out channels and canyons in a way that harder rocks would have been more resistant to. Yet the fact that Bretz couldn’t identify the source of the floodwaters caused many to dismiss his idea, and it wasn’t until evidence of similar events—such as Ice Age flood beds found in Montana—that other experts began to reconsider what Bretz had proposed. Eventually, by the 1970s, dismissive geologists changed their tune. Something catastrophic truly did transpire to create the Scablands.

Not that the behavior or history of such floods are completely understood. “There are loads of outstanding questions and many people are thinking carefully about the Scablands,” says University of Washington geology PhD candidate Kelsay Stanton. Even though experts are confident that vast glacial lakes provided the water for the floods, the precise volumes of the repeated floods are unknown, and the timing of the dozens of outbursts has yet to be determined in detail. “The glacial outburst flooding of the Pacific Northwest is hardly a closed subject,” Stanton says.

Massive canyons, like this one in Washington, were likely formed by powerful floods that repeatedly spread over the Pacific Northwest. Wolfgang Kaehler / LightRocket via Getty Images

Part of what’s allowing geologists to keep going back to the Scablands is that the tools available to scientists have changed a great deal since Bretz’s time. “There are lots of geochronology and computer modeling methods now that weren’t available when Bretz and other early researchers were mapping the area,” says University of Massachusetts Amherst geology PhD candidate Karin Lehnigk. These repeated floods affected the ocean, she notes, with the influx of freshwater reducing the saltiness of the northern Pacific for years and altering the way colder, saltier water in deep ocean layers circulated. The floods affected more than the land they ran over, and have acted as models for how our modern glaciers might alter ocean circulation as they melt due to global warming. New, broad scale methodologies can offer some of these insights and these refined techniques have proved to be critical, as no one has ever witnessed a flood quite like the ones that created the Scablands. “There aren’t modern examples of flooding on this scale that we can directly observe to figure out what processes are actually going on in the same way that people who study, say, rivers can,” Lehnigk says.

Work by Lehnigk and colleagues may even be able to reconstruct what the landscape was like before the dozens of floods created the Scablands. “I was doing fieldwork in 2017 with my PhD advisor and two of our colleagues,” Lehnigk recalls, “and we started talking about whether you could extend the pre-flood river profiles down in the canyon to rebuild the pre-flood floor.” That set Lehnigk on a path to look for clues on the modern Scablands about what the area was like in the Ice Age. Even though there’s no way to definitively recreate what the landscape looked like, she notes, the way water carves bedrock can be used to start with the current topography and work backwards to estimate what the starting conditions were like and detect variables like where rock was more likely to erode or be resistant. It’s a way to replay floods that we can no longer witness.

New techniques, including LiDAR scans to map terrain and numerical models used to estimate water discharged by glaciers, are turning up more evidence for various outburst floods in other time periods and places from the Mississippi River to the Himalayas—and even Mars. Explorations of the Red Planet have found landscapes very similar to the Scablands and other flood-created channels. Even before NASA missions found evidence of liquid water on Mars, geologic scars left no doubt that water once rushed over the planet’s surface. The research is reciprocal, helping researchers refine what we understand how such events occurred through space and time, a marriage of the old uniformitarianism with an understanding that change can occur quick and at scales totally unfamiliar to us. “It’s a two-way street,” Lehnigk says, “by learning more about the surface of Mars, we gain greater insight into surface processes on Earth.”

Get the latest Science stories in your inbox.