Lake Agassiz: New book explores origins, discovery of giant phantom lake

The book "Lake Agassiz: The Rise and Demise of the World's Greatest Lake" takes readers back thousands of years to when the lake formed, and up to the scientists and explorers who discovered it.

Evidence of lake's existence hidden in plain sight for millennia

Lake Agassiz spread over most of Manitoba, parts of Saskatchewan and Ontario, North and South Dakota, and Minnesota. (Submitted)

Bill Redekop drove from near the Canada-U.S. border south of Morden up to Swan River, following one of giant phantom beaches of the ancient Lake Agassiz.

"There were moments there that I thought were just stunning. It's 10 metres high and you drive right alongside it on some of these roads and it's really something to imagine," he said in an interview with CBC's Weekend Morning Show.

Known as Campbell Beach, it forms part of the hidden history of Manitoba, which he set out to explore in his new book Lake Agassiz: The Rise and Demise of the World's Greatest Lake.

Lake Agassiz was an enormous glacial lake that covered a large chunk of the North American landscape between 14,000 and 8,000 years ago, at the end of the last ice age.

Redekop's book takes readers back to that time, when life began returning to the landscape after being buried under ice for 50,000 to 75,000 years, and up to the scientists and explorers who discovered it.

For millennia, the evidence of its existence remained hidden in plain sight, but slowly details of the landscape began to merge in the minds of people passing through the province.

One of the earliest people to notice that evidence was a man from Philadelphia named William Keating, who travelled through the area that would become Manitoba in 1823.

"And he said at that time that it looked like there had been a giant lake here at one time," Redekop said. "And now in 1823, there was no real reference for a giant lake other than Noah's flood, so that would have been the only thing he might have thought, he could attribute it to."

Eventually the theory of an ice age began to emerge among scientific circles in Europe, but it wasn't until the mid- to late 19th century when scientists began to get a clear picture of the lake.

"They were starting to think differently about this, that there was a possibility that there had been not just mountain glaciers, but actual land glaciers, which no one had ever seen, so the evidence was all circumstantial."

Lake Agassiz spread over most of Manitoba, parts of Saskatchewan and Ontario, North and South Dakota, and Minnesota. (Submitted)

The lake changed its shape over the centuries, spreading over most of Manitoba, parts of Saskatchewan and Ontario, North and South Dakota, and Minnesota. The remnants of it can be seen in the three big lakes of Manitoba: Lake Winnipeg, Lake Manitoba, and Lake Winnipegosis.

"One beach was halfway through Brandon, so your east side of Brandon would have been in Lake Agassiz and the west side would have been beach-front property."

Redekop says writing the book has changed his view of Manitoba's landscape. 

"Now I have two landscapes: the one that I see and the one that I imagine, that I know was there 10,000 years ago."

Redekop will hold a book launch on Wednesday at McNally Robinson at 7 p.m.

Lake Agassiz: The Rise and Demise of the World's Greatest Lake 

Chapter 13


It's an extraordinary feeling to stand on one of Lake Agassiz's former beaches, even though many are fairly unimpressive topographical features.

The first time I discovered this, it came as a surprise. I didn't realize that I was standing on an ancient beach until someone told me. That particular beach was trifling in size, less than two metres high, perhaps 30 metres long and gently crested in the centre. It was probably the smallest beach I have seen. But this was my first Lake Agassiz beach, and it marked the first time I had knowingly stood on a prehistoric shoreline, and gazed out on ... dry land, and very flat dry land at that.

Yet to stand on one of these ancient beaches—or paleo- shorelines, as they are properly called—is to imagine standing along the edge of one of the largest lakes that ever existed. It is to hear the water murmuring on the shoreline on still days, and the waves crashing and showering you on stormy ones. I was on a southeast shore of Lake Agassiz, which would have been frequently pounded by waves from the prevailing northwesterly winds.

From where I stood, I could imagine an endless horizon of water stretching hundreds of kilometres to the west; a soaring two-kilometre high cliff of glacial ice to the north; icebergs, calved from the ice cliff, dotting the open water like sailboats; a deep, blue, birdless sky above; and the knowledge of how short my life would be in that imperfect world.

Someone did that. Someone stood there, maybe not precisely on that spot, but somewhere along the shoreline of this massive lake, and saw all of that, with his or her eyes, and his or her own thoughts.

These ancient dryland beaches are all around us but one would hardly know it today. No one speaks about them, or writes about them, or points them out. Only a few have signage, and only one has a highway sign, thanks to a group of diehard history buffs in the village of Arden. Good for them.

Yet it needs not be this way. Lake Agassiz's beaches are close enough together, running parallel and generally north-south, that the occasional roadside sign could easily clue people in. This is not to berate provincial heritage officials. It's just an idea that's long overdue. Manitobans are ready to learn about our incomparable glacial history.

This is, after all, the centre of the Ice Age.

Manitoba was central to the Wisconsinan glaciation, and home to one of the largest lakes the world has ever known. You'd think it would be hard NOT to know the locations of some of its former beaches. With ancient beach ridges all around us, you'd think people could point them out ad nauseam as they drive by. You'd think our kids would be bored to tears from all the times their parents had showed off how clever they were by indicating some wee blip on the horizon and declaring it "a paleoshoreline".

Alas, that's not the case. We don't know. We haven't arrived at an appreciation of our geological history yet and the colossal tale it tells. Yet Lake Agassiz's beaches are hiding in plain sight and practically doing star jumps and waving at us to be noticed.

Warren Upham summed up his findings of Lake Agassiz beaches this way: "The shorelines of Lake Agassiz are inconspicuous, though they are very distinctly traceable. They are usually marked by a deposit of beach gravel and sand, forming continuous, smoothly rounded ridges, such as is found along the shores of the ocean or of our great lakes.

"The beaches of Lake Agassiz commonly rise three to 10 feet [one to three metres] above the adjoining land on the side that was away from the lake, and 10 to 20 feet [three to seven metres] above the adjoining land on the side where the lake lay. In breadth, these beach ridges vary from 10 to 25 or 30 rods [from 50 to 150 metres]. The beach ridge is thus a broad wave-like swell, with a smooth gracefully rounded surface."

Part of the problem is that what we see in Manitoba are mostly just snippets of beaches. Lake Agassiz beaches were once fairly continuous shorelines, but not all formed discernable ridges. Lake currents prevented their formation in some places, some were later cut down by rivers and streams, and many have been grassed or forested over, making them hard to trace. What we see today are more like broken lines on a map.

This map shows the various shorelines of Lake Agassiz. (Submitted)

Upham still managed to map the beaches, without use of modern tools like aerial photos, Google Earth, and LIDAR (Light Detection and Ranging Imagery), which is similar to radar but uses laser light to more precisely map elevation changes. Where lengths of beaches had no rise, Upham read them by their sediments, which was a more reliable clue than a rise of the land. "The material of the beach ridge is remarkably in contrast with this adjoining and underlying till, for it includes no clay, but consists of stratified sand and gravel, the largest pebbles being usually from two or three to six inches in diameter," he wrote. 

What Upham was saying is that Lake Agassiz didn't leave ridges everywhere. Where they were missing, he had to map the sediments they left behind. A big clue was the absence of clay. Beaches were comprised of stratified sand and gravel, with clay mixes resuming on either side.

I would never have found Lake Agassiz's ancient beaches on my own. I was very fortunate to have former provincial geologist, Gaywood Matile, lend his expertise. I tried on many occasions to find my own beaches, using the old maps of Upham and W. A. Johnston, only to be told on virtually every occasion that a ridge I thought was a Lake Agassiz beach was created by something else. And perhaps that's the problem. Numerous land features mimic the look of ancient beaches. For example, I was fooled several times into thinking drumlins were beach ridges. Drumlins are ridges of glacial deposits left behind, similar to moraines. The difference is that moraines generally formed parallel to the glaciers, running in an east-west direction, while drumlins generally bisect the ice, running in a north-south direction. 

The following chapter includes some of the best spots in Manitoba to view beach ridges. These locations show large and very obvious beaches; in short, they are beaches that are worth a look.

As well, in the following chapter I have outlined the trip I took following the Upper Campbell Beach, one of the greatest examples of a glacial lake beach in the world. Over a three-day journey, I tracked its western shoreline, starting at the international border and continuing all the way to Swan River, 400 kilometres north as the crow flies. The Campbell Beach continues beyond that point, into Central Saskatchewan.

Lake Agassiz beaches have smoothly rounded ridges. They were created by waves piling sand and gravel, materials left behind by the glaciers. Wave action acts like the pan of a gold prospector, winnowing sand and gravel from other materials to form the beaches. The finer sediments, the dark grey clays and yellowish silts, settled in the lake's interior.

There are a total of 55 documented Lake Agassiz beaches representing 55 different lake levels. But many of the beaches are almost impossible to discern at ground level and are only visible in aerial photographs. These have not been named.

There are about 20 main beaches—those that are better formed than others—that once ringed the ancient lake, and they are named. The first three beaches—the Herman, Norcross and Tintah—were established while

Lake Agassiz drained south via the Minnesota River to the Mississippi River and out into the Gulf of Mexico. 
The Herman is the oldest, formed about 14,300 years ago, and the highest, reaching 320 metres above sea level at Lake Traverse, along the Minnesota-South Dakota border. Lake Agassiz fell an estimated 15 metres or nearly 50 feet over the next 1,000 years.

The Herman beach extended west all the way to Brandon, Manitoba's second largest city. Warren Upham maintained the Herman beach is still visible in Brandon, with a bay extending near Lorne Avenue, between First and Twelfth Streets.

That's debatable.

In Brandon, there is "a well-defined ridge of sand and gravel along a distance of about a mile," Upham wrote. "It extends from east to west, passing an eighth of a mile north of the court house, and thence close along the south side of Lorne Avenue from First to Fourth Streets. Between Fourth and Sixth Streets it is crossed by this avenue, and thence westward lies close on its north side. Its structure is shown by sections where it is intersected by Tenth, Eleventh and Twelfth Streets, exposing a thickness of 10 feet of obliquely bedded sand and gravel ... This beach ridge varies from 10 to 20 rods [50 to 100 metres] in width and from five to 10 feet or more in height, having a smoothly rounded wave-like form."
I walked this stretch a couple of times. It seemed to me to be jumbled terrain, but I did see what looked to be a ridge. The late Brandon University geologist John Welsted agreed with Upham that a bay from Lake Agassiz curved into Brandon, and wrote about it in 1988 in Brandon: Geographical Perspectives on the Wheat City. Welsted worked just down the street at the university and must have eyeballed the Herman beach numerous times.

But more recent opinion is skeptical. The sand and gravel ridge can be blamed on the glacial Assiniboine River that poured into Lake Agassiz, spewing sediments in a huge fan. That torrent of sediment-laden water, which formed the sprawling Assiniboine Delta, should have erased any glacial beaches inside Brandon, and the Herman Beach and geographically, the Assiniboine Delta occurred relatively close together.

After the Herman, Norcross and Tintah beaches came the Campbell Beach. Most paleoshorelines are fragmented and in low relief. The Campbell Beach is the exception. It looms large, certainly along Lake Agassiz's former western shoreline, and is virtually intact. It is by far the most impressive of Lake Agassiz's beaches.

"The Campbell Beach is unique. It's the only shoreline that you can even get close to tracing all around the lake," explained Matile. Where most Agassiz beach ridges are small, perhaps 1.5 to three metres high, and perhaps 100 metres wide, the depth of sand and gravel on the Campbell Beach is 10 metres or 35 feet high in places.

The story of its formation is also impressive. Imagine a lake being suddenly drawn down by as much as 100 metres, as happened on Lake Agassiz. On the great lake's eastern edge 12,800 years ago, a very low outlet is believed to have opened, channeling under a huge dam of ice and drilling through rock, abruptly draining Lake Agassiz through Thunder Bay.

Picture draining a depth of 100 metres of water out  of a lake, or out of the ocean or out of anything, in no time at all, geologically speaking. All around it, land that had been submerged for thousands of years was suddenly exposed. Some of it became dry land and plants quickly moved in to establish themselves on the sandy shores and the rich lakebed. Elsewhere, wetlands and marshes formed.

The land was exposed for between 1,000 and 1,500 years. Lichens, mosses and grasses were replaced by bushes and trees; forests grew up. Animals and human hunters arrived.

Then the eastern outlet became blocked again, likely due to a combination of readvancing glaciers and isostatic rebound. Lake Agassiz refilled, not all the way to the Tintah beach stage, but the water slowly reclaimed most of the dry land, drowning the marshes, grasslands and forests. And the beach that marks this remarkable return is the Campbell Beach. (Other, short-lived beaches undoubtedly formed as Lake Agassiz refilled, but all were subsequently erased as waters overtopped them.) Then, because beaches mark places where water levels remain constant for a number of years, we know the water level stabilized and Lake Agassiz's most prominent beach began to grow.
Upham didn't get around to naming beaches after Manitoba towns until later. So the newer beaches sport Manitoba names. They include Gladstone, Burnside, Ossowo, Stonewall, The Pas, Gimli, Grand Rapids and Niverville beaches. However, Niverville is suspected of being a ridge created by the keels of icebergs floating in Lake Agassiz, rather than an actual beach. 

These beaches were all created more than 4,000 years after the Herman. The oldest of the Manitoba beaches, the Gladstone, runs just east of the Town of Gladstone.

Just below the Gladstone Beach is the Burnside Beach, named for a village that once existed near the junction of the Trans-Canada Highway and Highway 16, and marking the place where Lake Agassiz's western shore lay. Farther east, the Burnside Beach once wrapped around Birds Hill, during a period when Birds Hill was an island (see page 125). It is the highest of the four beaches that ring Birds Hill.

Several beach ridges run parallel to the eastern boundary of Riding Mountain National Park. The park has a hiking trail, called Beach Ridges Trail, on its northeast corner. The beaches have been eroded and covered with vegetation, however, making them difficult to detect. The way to find them is by their trees. The dry, sandy, better-drained soils of the former beaches are preferred by stands of bur oak, and there are swathes of oaks on Riding Mountain's eastern slope. Once the forest along the trail changes to poplar, you know you're off the beach.

There are two beaches on the Hudson Bay Railway, on miles 109 and 110 from The Pas, believed to be related to the Grand Rapids Beach, one of the lowest of Lake Agassiz's beaches. The last two beaches before it drained into Hudson Bay are the Kinojevis Beach—Kinojevis being a river in eastern Quebec—and the Fidler Beach, named for fur trader and surveyor Peter Fidler. The Kinojevis Beach was formed when glacial lakes Agassiz and Ojibway combined, while Fidler was the final beach before the waters flowed into Hudson Bay, or its larger ancestor, the Tyrrell Sea.

Matile and I drove southeast of Steinbach on Highway 210. Just before you get to the junction of highways 210 and 404, there is a small, unimpressive beach ridge. It can't be more than 30 metres long, and perhaps two metres high. Highway 210 runs through it. It has its own marker, the twin road signs for Highways 210 and 404 fastened to a post. It just takes a couple of minutes to walk its width and feel connected with Manitoba's ancient history.

Another well-marked beach ridge is on Highway 12 at the Saints Peter and Paul Ukrainian Orthodox Cemetery, just north of Sarto. If you look across the road from the Ukrainian Orthodox cemetery, you see falling-down fence posts following a small arc in the landscape. That's the continuation of an Agassiz beach. Matile pointed out similar small beaches ridges along Highway 12. (The directions in Chapter 14 are for much more prominent beaches than these, with a definite wow factor.)

In fact, one of the best locations to see Lake Agassiz beaches is right around Steinbach. The reason for this is Steinbach is at the edge of Lake Agassiz's clay plain, where the land starts rising to the east into the Canadian Shield. The rise means the lake beaches are clustered together. 

The main beach Matile took me to see was the eastern Campbell Beach at the base of the Sandilands Provincial Forest. The Sandilands were submerged during Lake Agassiz's deepest stages but later became a prominent island. So the Sandilands isn't so much a beach ridge as the edge of an island that the Campbell Beach wrapped around. Standing on a sand cliff in the Sandilands facing west, Matile said we were looking across more than 300 kilometres of Lake Agassiz. The shore across to the west would have reached to Brandon. That was quite a lake.

Sandilands Provincial Forest is part of an extraordinary land formation that we don't fully appreciate except where it turns into Grand Beach, one of the world's best beaches. We can thank our glacial past for that.

Sandilands has been thought of by some as an end moraine—a place where the ice sheet stalled. That means the glacier was melting and refreezing in lock step, allowing debris to drop off its leading edge for perhaps hundreds of years and forming a ridge.

Today, however, this land formation is believed to have been created by what's called outwash, and that's Matile's interpretation, too. Outwash is runoff from glaciers. This particular outwash formation was at the juncture of two enormous glaciers: the Labradoran glacier with its dome east of Hudson Bay, in northern Quebec; and the Keewatin glacier, with its ice dome west of Hudson Bay, in Nunavut.

The Keewatin glacier was ploughing south, and the Labradoran glacier was ploughing southwest, and eventually they ran into each other, or at least side-swiped one another, to form one giant ice mass we know as the Laurentide Ice Sheet. But when they started to melt, or become "unzippered", as Matile put it, the sand that had accumulated inside the glaciers was washed by meltwater into the giant trough opening between them. The accumulation from that trough is the vast raised expanse of sand we see today, stretching from Elk Island in Lake Winnipeg just north of Victoria Beach through Grand Beach, Belair Provincial Forest, Mars Hill Wildlife Management Area, Milner Ridge, Agassiz Provincial Forest, and Sandilands Provincial Forest to the international border.

This raised mass of sand is 250 kilometres long, and it's often mistaken for a moraine. But moraines are made of till—the geological name for a mix of sand, gravel, silt and clay. This extensive ridge is too sandy to be a moraine, Matile maintains, and the sand is 30 metres deep. It had to have been formed by outwash.

The Sandilands are the highest point in southeastern Manitoba, reaching an elevation of 366 metres above sea level. The Campbell Beach surrounded them but near the bottom. While Campbell Beach was at the foot of the Sandilands, one of the older beaches, the Norcross Beach, crossed higher up on the Sandilands.

Adding to the complexity of studying the Sandilands is that the prevailing winds were northwesterly, or nor'westers. These sent giant waves crashing into the Sandilands, eroding sand on points that were sticking out into Lake Agassiz. The waves carried the sand along the shoreline to the south, forming spits, not unlike sandy ocean spits. Elaborate beaches were developed this way; they are now covered with trees and other vegetation. (Other complexities in these ancient beaches are that some are wave-cut scarps, where wave action has eaten into small shoreline cliffs. The lake also formed offshore sand bars leading up to a beach, and experts often can't tell the difference between beaches and sand bars.)

The Campbell Beach would have skirted the west side of Whitemouth Lake in southeastern Manitoba, explained Matile. It might even have acted as a barricade that contained the water and helped to create Whitemouth Lake.

From the Sandilands, Lake Agassiz's beaches get lost in the Canadian Shield, for the enormous lake stretched as far as Rainy Lake and Sioux Lookout in northwestern Ontario. 

If you're wondering why there are no beach ridges in the Canadian Shield, it's because the Shield was an archipelago of islands in Lake Agassiz. The rock ridges poking out of the water muted the power of Lake Agassiz's wave action and prevented the development of good beaches inside the Canadian Shield.

So, who cares? What difference does it make that Manitoba has ridges of sand and gravel running all over the province? The reality is that we all cared. The early Indigenous people, the settlers, the first highway and railroad builders, and the grave diggers, all cared. Even animals cared, and often followed beach ridges when they migrated or moved from one area to another.

Lake Agassiz's beach ridges made awesome trails for Indigenous people beginning with the lake's earliest paleoshorelines. Much later, they did the same for fur traders, and European settlers with their ox carts and horse-drawn wagons, particularly when the travel was north and south. The beaches hindered east-west travel somewhat because their north-south configurations trapped water behind them, creating lagoons and sloughs.

In the spring, when the grasslands and woodlands were wet, people could travel and hunt on the beach ridges. Because of their sandy soil and elevation above their surroundings, they were the first to dry. Also, forests were unable to thrive as they could elsewhere. Only bur oaks could stand the sand and swift drainage. As Joseph Tyrrell wrote in the 19th century, "They often form beautiful dry roads through the country that would otherwise be an impenetrable forest."2
Beach ridges also served as lookouts for hunters spotting game or potential enemies. They made good camp sites because they were high and well-drained, especially where streams cut through a beach ridge and provided water. 
An added bonus for early peoples was the quartz or Swan River chert often found embedded in beach deposits. Chert, prized for making spear and arrow points, is found in large concentrations in Lake Agassiz beaches. 

The beaches also made natural burial grounds, both then and today, because the ground was sandy and water drained away instead of ponding. It was also easier to dig graves in the sandy soil, especially before metal tools were routinely used. Both native North Americans and Europeans used them for cemeteries. There are dozens of cemeteries on top of beach ridges in Manitoba. In fact, that's one way to spot a beach ridge.

For all these reasons, beach ridges have great cultural significance as archaeological sites and First Nation burial grounds. Anthony Buchner, a senior archaeologist formerly with the Province of Manitoba, observed that the beaches were used by Indigenous people for religious or ceremonial activities. Manitoba's only remaining medicine wheel, a circle of stones with spokes, near Alonsa on the west side of Lake Manitoba, is on a beach ridge. "The medicine wheel served a variety of functions including spiritual cleansing, and appeasement of mystic powers and beings," wrote Buckner.

Many beaches became roadbeds for railways and highways for the same reasons that they were used by earlier Indigenous people and settlers. They were elevated and well-drained. Railway companies and later road construction contractors could save money by building on top of them.

For example, Highway 308 from East Braintree to Moose Lake and Sprague follows an old Lake Agassiz shoreline. Highway 10 follows Campbell Beach between the TransCanada Highway and Cowan. Highway 352 also follows Campbell Beach. Highway 50 east of Lake Manitoba follows the Burnside Beach, which holds back drainage into Lake Manitoba, allowing the formation of the Big Grass Marsh near Gladstone.

Beaches created a tremendous sand and gravel industry for Manitoba. The former Agassiz beaches have provided materials for roads, railroad ballast, and aggregate for building construction. I searched and followed many Lake Agassizbeaches and there was a sand and gravel pit somewhere on each one. Some pits were quite small, and some were massive. Lake Agassiz separated the sand and gravel from the silt and clay to allow the quarrying. There are 3,000 provincially owned and operated sand and gravel pits across Manitoba and perhaps twice that number of private pits, and Lake Agassiz is a major reason why.

Near Beausejour, Lake Agassiz produced fine sands, with a high silica content. As a result, the first glass container factory in Western Canada was established there in 1906. Manitoba Glass Works employed glass blowers from Poland and the United States, assisted by local employees, to make bottles for breweries and soft drink companies in Winnipeg.

These were glass bottles with the greenish hue; examples can often be seen in old bottle collections. The green is not an indication that the glass is aging, but the original colour. Manitoba Glass Works later produced jars, and medicine and ink bottles, and employed up to 350 people. But it was short-lived. Competition from larger manufacturers in Eastern Canada forced it to close in 1914.

With so many different beaches, I had to ask Gaywood Matile a question. Was Lake Agassiz really a lake? Lakes are supposed to be stationary, right? They aren't usually shape-shifting water bodies, constantly rearranging their boundaries. That sounds more like a long 6,000-year-old flood. 

"You're thinking human time versus geological time," Matile responded. "Lake Winnipeg is moving, too but you don't really see it. Isostatic rebound is still at work. That's why you have a drowning south shoreline."

With files from the Weekend Morning Show