Support Provided ByLearn More
Planet EarthPlanet Earth

Megafloods of the Ice Age

About 15,000 years ago, in the waning millennia of the Ice Age, a vast lake known as Glacial Lake Missoula suddenly burst through the ice dam that plugged it at one end. In the space of just 48 hours, geologists believe, the collapse sent 500 cubic miles of water cascading across the Pacific Northwest, creating overnight such unusual landscapes as the scablands of eastern Washington State. In this interview, geologist and scablands expert Vic Baker of the University of Arizona talks about what it would have been like to witness the flood, what such floods are teaching us about Mars, and what J Harlen Bretz, the colorful geologist who first proposed that cataclysmic flooding had carved the scablands, was really like.


Recieve emails about upcoming NOVA programs and related content, as well as featured reporting about current events through a science lens.

Scores of times during the Pleistocene epoch, Baker says, floodwaters from Glacial Lake Missoula (lower right in image) punched through the ice dam, inundating vast swaths of the Pacific Northwest.
Support Provided ByLearn More
Created to scale with ArcView and Visual Nature Studio by Rick Lovel, HDR Engineering, Boise Idaho.

In the path of the floods

NOVA: What would it have been like to observe the floods from a safe vantage point?

Vic Baker: Pretty frightening. A lot of water was transferring very quickly to the land surface, so this would probably have caused some vibration. Maybe not full-scale earthquakes, but depending on how far away you were, there would have been a lot of stress and a lot of noise from boulders banging together. You might also have seen rather bizarre things—huge waves, for instance, perhaps even particles flying out of the flow. That would have been rather disconcerting.

You mean like an explosion of particles?

I don't know if it's an explosion. It's just that there are turbulent forces with tremendous lift. You can think of it like tornadoes that lift large objects into the air. This would happen in the water, and if the momentum was sufficient, particles could be thrown out of the water. Certainly the Missoula flows were of the magnitude that that sort of thing could have happened.

Any chance there were people around at the time who could have witnessed them?

There's no direct evidence that people were impacted by the flood. The flood occurred prior to the major evidence for the first large group of prehistoric Americans, the so-called Clovis culture of big-game hunters. Clovis occurred at least a few thousand years after the flood. But it's possible people were there.

If there were people, they would have been swept away, of course. We know that large animals were taken up by the flooding. Vertebrate bones of various late Pleistocene animals, including mammoths, have been found in flood slackwater deposits. If they were around, people would have been similarly inundated. But nothing has been found.

Kolks, a kind of underwater tornado that occurs in deep flows of fast-moving water, tore huge potholes in the scablands and probably also flung debris right out of the floodwaters.
© Mentorn/WGBH Educational Foundation, animation by Darkside @ Molinare

You said "flood." I read an article recently claiming that there's evidence for only one Glacial Lake Missoula flood or perhaps several. Where do most geologists stand on this?

The evidence is overwhelming that there were multiple floods. Actually, the issue isn't so much were there multiple floods, but how big were the floods in the late Pleistocene? There's pretty good evidence that we're talking about scores of floods, maybe 100 or more. But it's less clear how big each was. There's a fair bit of evidence that many of them were extremely large by modern standards, but many were much smaller than others.

"Portland, Oregon would be under several hundred feet of water."

The number of really big floods is important because we think that in order to do significant erosion and deposition—at least in the main channels, not in the slackwater areas—you have to have a very high threshold for the activity. And therefore, the question is more, Did one flood (or a small number of large floods) in the late Pleistocene do most of the work? Or was this the accumulated work of about 100 much smaller events? We still don't know the answer to that question.

If one of the biggies happened today, what would be lost? Spokane, for instance, lies right in the path of the floods, right?

Well, it's not just Spokane. Portland, Oregon would be under several hundred feet of water. The whole system along the Columbia River would be affected tremendously by the flow.

But there hasn't been a lot of interest in what would be lost other than the Hanford nuclear site. Since they had put a whole lot of nuclear technology and radioactive materials right in one of the major confluent zones of Missoula flooding, they actually did, 20 years ago, some studies of the effects Missoula floods might have there. It was an academic exercise, of course, because it's not likely this would occur again unless there was an ice age. And an ice age would cause a lot of problems besides just nuclear waste!

Icelandic glaciers like this one have released damaging floods, but nothing on the scale of those generated by Glacial Lake Missoula.
© Mentorn/WGBH Educational Foundation

Megafloods around the world

Are there any other places in the world where such massive glacial floods could happen today?

Nothing of the magnitude that occurred in the late Pleistocene. The Glacial Lake Missoula flooding—and other floods we know about that occurred around the same time in geological history—were associated with extremely large ice sheets. When you have large ice sheets, you have not just the water in the ice but big lakes that those ice sheets dam up. So there's a lot of water available for producing the volume that such megafloods would entail.

Today, we only have two big ice sheets on Earth, and both of them almost completely fill landmasses—Greenland and Antarctica. So you don't have so-called ice marginal lakes, though in Antarctica you do have some subglacial lakes.

That brings up another issue: could big subglacial lakes release floods? Doesn't look like that's the case today, except with much smaller glaciers like those in Iceland. There you can get subglacial lakes, particularly associated with volcanism, that release floods. But neither the glaciers nor the lakes are nearly as big as anything we saw in the past.

"One reason I study the Channeled Scabland is that it's helpful in understanding Mars."

You mentioned other great floods in the Pleistocene. Where else have multiple megafloods taken place?

We had huge floods around many of the big ice sheets, such as the Laurentide, which blanketed much of arctic Canada east of the Rocky Mountains and southward to the Great Lakes. Huge megafloods were associated with its surrounding lakes, particularly Glacial Lake Agassiz, which covered much of central Canada centered around Winnipeg. That released floods both to the south through the Mississippi River system and also under the ice out through the straits that enter Hudson Bay.

There were also massive megafloods in Asia. Some of them were associated with big ice sheets that blocked the rivers that currently flow north from Russia into the Arctic Ocean. Others were associated with mountain areas, such as the Altai Mountains, the Sayan Mountains, and some of the mountains around Lake Baikal. All of these are in southern Siberian Russia, along its borders with Kazakhstan and China and Mongolia. These mountain floods were comparable in magnitude, though perhaps not in volume, to the Missoula floods.

Billions of years ago, Mars hosted megafloods that dwarfed those that formed the scablands.
Courtesy NASA

Landscapes of mystery

Getting back to the scablands, are there geologic features there that still puzzle geologists as to their origin? Or have most of them been figured out?

There are always puzzling things. For one thing, many of the processes that occurred during the cataclysmic flooding are of a scale that we can't reproduce in the laboratory, and they're difficult to evaluate theoretically because of not having that kind of experimental test. There are features of erosion and sediment transport, for instance, that we don't fully understand. We have hypotheses for most things we see, but we still have to see how effective these hypotheses are.

We try to advance this by finding other places and other examples where we can get some kind of constraint on how those processes operated. So these other megaflood areas I mentioned are important in helping us understand cataclysmic glacial breakout floods as a general phenomenon, not something unique, say, to Missoula.

Are there other entire landscapes that stump geologists as to their origin, as Washington's Channeled Scabland once did?

We're finding them all the time. Most interesting and puzzling are those we're finding on other planets. We know from work over the past 30 years that Mars, for instance, has had even bigger floods on its surface than we see in the Channeled Scabland. Those floods seem to have been both similar to and somewhat different from the Missoula floods. In many cases they involved much more water and even larger flood-eroded landscapes than in the Channeled Scabland.

We could spend this whole interview talking about Mars, because there's so much that is not fully understood there, and we only have incomplete information about it. One reason I study the Channeled Scabland is that it's helpful in understanding Mars as well as what we might find on other planets in the future, not just in our own solar system but beyond.

Ice sheets
"Some of the most puzzling landscapes are those that developed beneath the big ice sheets," Baker says. Here, a rendering of the Cordilleran ice sheet, whose Purcell lobe periodically dammed up the Clark Flork River, creating Glacial Lake Missoula.
Created to scale with ArcView and Visual Nature Studio by Rick Lovel, HDR Engineering, Boise Idaho.

What about here on Earth? Are there whole landscapes that still leave geologists scratching their heads?

Yes, and they're in surprising places. I think some of the most puzzling landscapes are those that developed beneath the big ice sheets. We have a few such landscapes today under Antarctica and Greenland. Also puzzling are ones that were exhumed from beneath the big ice sheets in British Columbia, which was covered by the Cordilleran ice sheet, and in much of eastern North America where the big Laurentide ice sheet was.

Those landscapes are not very well understood because we don't really have a modern analogue to them. In geology, we try to understand things to a large extent by a sophisticated use of analogy, where we have features that are being formed by processes today or in some way that we can get a handle on how those processes operated long ago. Where we understand the relationships and then we try to extrapolate that into unknown landscapes.

"Some people think science is the collection of facts and truths and everything about the world. Absolutely not."

The big ice sheets of the last ice age were in many cases warm-based ice sheets. That is, there was a lot of water underneath the glaciers. The landscapes that form there are a complex combination of ice and water processes involving meltwater and high pressures. There are all kinds of bizarre features on those landscapes that we don't fully understand. That's one category. I could go into other parts of the world, say in the tropics, where there are completely different situations going on that also pose puzzles.

I guess I was just wondering if there's still a lot remaining to be solved geologically at the landscape level, and from what you're saying there is.

Yeah, the fact that we have science today is by definition an indication we haven't solved things. This is an important point. Some people think science is the collection of facts and truths and everything about the world. Absolutely not. Science is about raising questions about the things we don't know and being very sophisticated about pursuing those problems. If everything was solved, there would be no science. There'd be discussion of facts, but there wouldn't be science.

J Harlen Bretz
J Harlen Bretz, whose first name was "J" (no period necessary)—just one of the many idiosyncrasies of this colorful geologist
Courtesy Special Collections Research Center, University of Chicago Library

The two who first knew

So you knew J Harlen Bretz. What was he like?

Oh, he was one of the real characters in geology. That's one of the great things about geology, as in science as a whole. We have our regular bookish-type people, but we also have what you might call characters—people who've got their warts but also their shining sides.

Bretz was a complete geologist. He was obsessed with his work, and he was true to wholly geological things, like multiple working hypotheses and looking in the field for evidence as opposed to theorizing from an armchair. He formed strong opinions. He loved working with students, but he pressed those students hard by making them think for themselves. He was a Socratic teacher, where you don't tell people, you let them figure it out for themselves. You're more like a guide in their process of understanding.

Bretz also enjoyed life. He liked making his own wine. He liked telling jokes. He liked going on family trips. But he was also an extremely difficult person to live with, I'm sure. He was tough on wife and family, and probably a bit full of himself. So he was remarkable in many ways.

Now, if Bretz had given solid evidence for catastrophic draining of Glacial Lake Missoula at that 1927 geological meeting, do you think geologists there would have accepted his theory? [Editor's note: At the meeting, Bretz presented his theory that a massive flood had formed the Channeled Scabland, though he admitted he had no idea where such a volume of water had come from; he was lambasted by other geologists that day and for years afterward.]

I don't think at that meeting that that evidence would have carried the day. But in subsequent reflection, if the evidence was as Joseph Pardee later presented it in the 1940s, then I think they would have to have come around to Bretz's view earlier. Because that was clearly a process that had occurred and was of the magnitude to have generated the flooding.

People have said geology's kind of like a murder mystery, where you sometimes get a smoking gun that indicates the culprit. By that view of geology—which I think is partially right but not completely—the Missoula breakout would be the smoking gun that helped solve the mystery. So yeah, it could definitely have had a major impact.

Joseph Pardee
Joseph Pardee suspected as early as the 1920s that Glacial Lake Missoula was the source of the floods that scoured out the scablands, but he didn't publish his theory until 1942.
© Geological Society of America

Why didn't Pardee speak up at that 1927 meeting if he knew then where the water to carve such landscapes had come from, namely, Glacial Lake Missoula?

Well, we don't know much from Pardee himself, because he was very quiet about that. Certainly Bretz had an extremely strong view on why that was the case. Bretz told me this himself, and he wrote about it. He felt that Pardee was prevented from speaking up. Pardee was an employee of the United States Geological Survey, and his superior was a fellow named Alden, who was extremely conservative and antagonistic to the flood hypothesis. Alden had been one of the critics at the meeting in Washington.

"He was happy to be alive, for one thing."

Thus Pardee was there as a junior person with other more senior USGS people, all of whom were against the flood hypothesis. So Pardee himself, being a rather quiet, taciturn person, was certainly not going to speak up at that meeting. Bretz felt—and this has not been completely verified by the paper trail—but Bretz felt from people he had talked to that Pardee had actually thought of cataclysmic flooding as a possibility for the landforms he observed around Spokane a few years prior to Bretz thinking of it.

That may be another reason Pardee didn't speak up: he saw Bretz getting in trouble for the exact thing that Pardee had thought of but hadn't been able to publish because of the conservative nature of his superiors. So it was kind of a combination of not wanting to go against his superiors and also being a little bit miffed that Bretz had come up with the same idea that he had had earlier.

Vic Baker
Vic Baker returns regularly to the bizarre landscapes known as the scablands, in part because of what they can tell us about similar terrain on Mars.
© Mentorn/WGBH Educational Foundation

Eventually, of course, Bretz was vindicated. Did he ever talk to you about how he felt when he finally was?

Oh, well, he was delighted about that. He didn't gloat about it to me, but clearly he was happy to be alive, for one thing. He had gotten through the process; he had outlived his critics; and he could see things come around to what he had been saying, because people finally went out in the field and saw what he was talking about. He was also happy to see that new stuff was happening, that people were building upon the train of thought he set in motion.

I think that's the kind of thing that we in science take satisfaction in—not that you have absolutely solved the problem, because then there's no science left. It's that you have helped in pointing the direction toward a productive line of work and are part of this larger human enterprise of discovery and creativity. I think he probably felt a bit of that. You can get self-centered in science, but I think to be a good scientist, you've got to break away from that. I think Bretz was able to do that.

Editor's Notes

This feature originally appeared on the site for the NOVA program Mystery of the Megaflood.