• Fraser Valley Current
  • Posts
  • The benefits, opportunities, costs and challenges of restoring Sumas Lake (Part 1)

The benefits, opportunities, costs and challenges of restoring Sumas Lake (Part 1)

Restoring Sumas Lake would bring uncalculable environmental benefits—and costs and flood challenges

This story first appeared in the June 11, 2024, edition of the Fraser Valley Current newsletter. Subscribe for free to get Fraser Valley news in your email every weekday morning.

Technical issues have forced us to split this story in two. You’ll find the link to the second part at the bottom. You should read this story in sequence. If you need to pause and would like to return, you can find links to the different sections of the story below.

There’s no cheap—or simple—option for reshaping the future of Sumas Prairie.

The City of Abbotsford has been denied billions of dollars to build new dikes and a pump station to protect the prairie against a future flood. Meanwhile, a new study suggests that buying out properties on the former bottom of Sumas Lake and allowing the lake to return would come with a significantly lower price tag.

The twin pieces of news sparked a rash of commentary from provincial politicians. BC Conservative Party president Aisha Estey suggested that buying out properties in a form of “managed retreat” might be the best long-term option for the future. That led BC United to distribute a press release vowing to oppose what it said was the Conservatives “plan to flood Sumas Lake.” BC’s Agriculture Minister and Abbotsford-Mission MLA NDP Pam Alexis also described the suggestion as “shocking.”

It wasn’t really all that shocking. The idea of a reincarnated Sumas Lake has been percolating ever since the 2021 flood. And yet, as conversations about the future of Sumas Prairie—or Sumas Lake—took prominence last week, most failed to recognize the complexity of the discussion.

Having decided that it wants to preserve a key economic driver of Abbotsford’s economy, the city and fellow backers of stronger flood protections have never undertaken a full accounting of the costs and benefits of restoring the lake.

But the same can be said of proponents of reestablishing Sumas Lake. They have omitted considerations about the financial and economic costs related to re-orienting transportation and energy infrastructure, and re-imaging BC’s agriculture economy. And in making their case that buyouts would be cheaper, they have put borders on a lake that never had any.

This is a long story. We get it. You can jump back to your place using these links.


The costs of protection

The benefits and costs of restoring Sumas Lake cannot be evaluated in isolation. Instead, they must be compared to both the status quo and proposals to reconfigure Sumas Prairie flood defences.

Fortunately, many of those comparators already exist. The 2021 disaster has shown these costs of floods—and not preventing them or attempting to mitigate the inevitable damages when they occur. 

The future mitigation costs are (somewhat) known: the City of Abbotsford has spent two years sketching out reconfigured flood protections—and putting a dollar figure on those plans.  The Abbotsford flood plan would use dikes to create a narrow floodway to funnel overflow waters from the Nooksack in a northeasterly route from the US/Canada border to the Fraser River. The path would move the water beneath Highway 1, along the base of Sumas Mountain, and through a new pump station at the northern part of the prairie. That pump station would be needed to move water out of Sumas Prairie and into the Fraser River when the Fraser is higher than the water in the prairie, as was the case in 2021. A dike would prevent the water from entering the basin that once held Sumas Lake. 

The City of Abbotsford’s planned flood protection works would cost as much as $3 billion. Click photo for a larger version. 🗺 City of Abbotsford

The new study published last week—which was jointly authored by UBC researchers, scientists from environmental advocacy organizations, and Semá:th First Nation councillor Murray Ned—says it would cost $2.4 billion for those dikes and the new pump station. In a radio interview, the president of the BC Conservatives referred to the study and suggested it demonstrated the practicality of restoring the lake. That comment triggered responses by BC United and Alexis. In fact, the $2.4 billion figure actually came from a previous City of Abbotsford analysis of possible options that predates the current plan. The current price tag for all flood protections—including new pump stations, reconfigured dikes, and some property buyouts along flood zones—is even higher: around $3 billion. Given the time that would be needed to undertake the work, inflation is likely to increase those costs in the years to come.

The largest unknown for flood protection concerns the eventual rebuilding of Highway 1 between Whatcom Road and the Vedder Canal. Not all of that highway project is related directly to protecting the route against flooding. But plans for that stretch of highway are still in the works and are likely to see the highway raised to protect against future flooding. Inevitably, creating a more resilient highway would add millions—probably hundreds of millions—to the project’s cost.

There are other costs too.

Long-term, flood protections like dikes and pump stations come with significant maintenance bills. Operating an additional Barrowtown Pump Station, maintaining dikes to an acceptable standard, and preventing erosion, sinkholes, and other natural wear-and-tear will add to the lifetime expense of flood protections.

Inevitably, reliance upon a flood protection system is accompanied by risk. As 2021 showed, natural events can cause dikes to fail, leaving residents and governments on the hook for billions of dollars of damage. Paying for a dike system does not guarantee that the system will work 100% of the time. And like all human infrastructure, dikes and pump stations won’t survive forever. Eventually they will crumble and need replacing. It’s also possible that climate change will render them redundant before that time comes. The designers of a new dike system will need to balance resiliency and preparations for future climates and storms with the increased costs that come with super-sizing flood protections.

There are other, less obvious costs too. Abbotsford’s current plan would involve redesigning dikes and adding pump stations to allow some overflow to accumulate in the Sumas Lake area during very large storms. (Doing so would allow the lakebed to function as a release valve when water rises towards the crest of dikes.) That process could lead to a degree of unwelcomed flooding on cropland. 

Building new dikes and creating floodways will also require buying land from existing property owners. That will come with a monetary price tag, but also potentially a social and cultural one for any landowners who need to give up their land to protect other farms. 

Finally, bolstering flood protections will also preserve both the present-day benefits and the costs of a food hub built on an artificial prairie. We’ll explore the ongoing natural, environmental, and social costs of the removal of the lake as we break down the benefits of allowing Sumas Lake to return. And we’ll discuss the social and economic benefits of the prairie’s current state when we discuss the costs of allowing Sumas Lake to return.

The lay of the land

Advocates for the restoration of Sumas Lake have spent the last two years making the case that a renewed lake would bring an array of significant benefits to the region. Returning the lake to the valley would be a form of “managed retreat,” a concept built around the idea that it can often be safer and more cost-effective over the long-term to move people out of disaster-prone areas rather than pay for ongoing protection against natural forces. Managed retreat emphasizes the well-founded belief it is usually easier to work with nature than against it.

But it’s important here to understand that Sumas Lake was not a typical lake—and that even before Europeans arrived, Sumas Prairie was, more than most, a place where geography and waterways were in constant flux. 

Both the lakebed and surrounding prairie are extremely flat, thanks to the silt that washed into the area when the Fraser River rose in semi-annual floods. The boundaries of Sumas Lake would shift continuously. As the river receded in late summer each year, the lake would contract dramatically. Much of the lake bed would become a broad wetland. When the Fraser rose—whether in late spring during freshet or in the winter following heavy storms—water would wash into the valley, creating an expansive lake. That lake usually occupied an area of around 9,000 acres in what today is the eastern Sumas Prairie. But when the water rose especially high, as it did in 1894, the lake could extend all the way to what is now Huntingdon and Sumas, Wash.

Occasionally, after large winter storms, water would arrive in the prairie, not from the Fraser, but from the Nooksack River to the south. There was even a time—perhaps within the last millennium—when the Nooksack River flowed north and contributed its water to Sumas Lake and the Fraser River. 

🗺 Tyler Olsen/Google

Restoring Sumas Lake would likely involve the removal of Barrowtown Pump Station, the key piece of machinery that keeps the Fraser River out of the prairie and prevents Sumas Lake from returning. Its removal would again tie water levels in a resurrected Sumas Lake directly to those in the Fraser River—and, to a lesser extent, the nearby Chilliwack/Vedder River. (The precise mechanism by which Sumas Lake would be restored has never been spelled out. It’s possible floodgates could be added to try to regulate the size of the lake. Those would bring their own complications.)

In 2021, Sumas Prairie was flooded largely because of overflow from the Nooksack River south of the border. Restoring Sumas Lake would not prevent future Nooksack River floods. A plan would still be needed to, at minimum, prepare for the arrival of flooding from the United States. The potential size and scope of Sumas Lake in the event of a Nooksack River flood would be dictated by a variety of factors, including the height of the Fraser and the Vedder rivers, the scale and duration of a Nooksack overflow, and the pre-flood size of the lake.


Flood protection

The restoration of Sumas Lake comes with a simple, appealing argument: that it is both safer and more cost-efficient to buy out current residents and businesses who would then relocate to less flood-prone areas. 

This is the chief disaster-prevention argument for a restoration of the lake. By moving people and property out of the way of future floods, billions of dollars of future damages can be avoided.

The 2021 flood killed hundreds of thousands of farm animals, destroyed barns and equipment, and ravaged crops. If the aggregate value—whether in agriculture products or other types of assets—had been located elsewhere, it would have survived. There were also human costs: the flood forced thousands of people to evacuate, and left many of those residents with long-term trauma. Future floods could do the same. Moving people out of the way of future floods could forestall human tragedies in the decades and centuries to come. (We’ll get to the complications of this, including the fact that much of the losses did not occur on the lakebed itself, later.)

There are additional disaster-avoidance benefits when it comes to restoring a lake and wetland. The region’s water systems are all interconnected, and restoring Sumas Lake would increase the total water capacity in natural spaces in the Lower Fraser River basin. Confining rivers within dikes can “squeeze” water during peak flow events, meaning it takes less water to increase the height of rivers. Significantly increasing the flood storage capacity of the valley would have the opposite effect: when rivers swell during spring freshets and following storms, Sumas Lake would allow the water to “spread out,” reducing the water’s overall height and flood danger.

Depending on the time of year, the lake could potentially also mitigate flooding during future Nooksack River overflows. (This is largely unstudied and would depend on the calibration of flood defences elsewhere in Sumas Prairie. It’s also possible that the existence of a lake could slow the northbound draining of water into the Fraser River and aggravate flooding in the Sumas and Huntingdon area. Reclaiming Sumas Lake would still require a plan and infrastructure to prepare—whether passively or actively—for the Nooksack River’s occasional devastating floods.)

The return of Sumas Lake would demand a philosophical rethink of how humans interact with waterways in Sumas Prairie and how one prepares for high-water events. As the collapse of the dike protecting the lakebed demonstrated, people, governments, and businesses can overrate the resiliency of physical infrastructure as protection. That’s especially true over the long-term, as memories of previous failures fade. Proponents of managed retreat note that by preventing regular and frequent floods, dikes encourage development in floodplains. Increased development and human activity lead to increased disruption and damage when flood infrastructure is incapable of doing its job. 

Most flood infrastructure in BC is created for events of a certain annual probability. (Frequently these are described as a one-in-X-year flood.) A one-in-500-year flood has only a 0.2% chance of occurring in any single year. But over time, those little chances add up, and even the strongest of dikes can fail in the force of random weather chance. The odds of major floods, especially those caused by storms, are predicted to get even worse as climate changes.

As climate change increases the potential for floods, removing people from their path may be more economical in many instances than spending millions improving and maintaining flood protections.

“Dyke rehabilitation programs tend to assume that future waterflows will be predictable,” Riley Finn, one of the authors of the UBC study is quoted as saying in a press release. “However climate projections show that flooding events are likely to increase in the future—and the water needs somewhere to go.”

The restoration of Sumas Lake and its fluctuating levels would force humans to accommodate, rather than try to control, the water that surrounds them.

*The consequences of climate change on the potential for Sumas Prairie flooding are complex. Climate change will increase sea levels, which has an impact on the height of the Fraser River. It’s also expected to increase the frequency of atmospheric rivers and the floods they cause on both the Nooksack and Fraser. On the other hand, by decreasing snow packs, warmer temperatures may lower the scale or frequency of major freshets.


The environmental case for allowing Sumas Lake to return has been articulated in depth by proponents, Indigenous communities, and historians.

Chad Reimer’s book, Before We Lost The Lake, provides perhaps the most comprehensive look at the lake’s role in the Fraser Valley’s interconnected ecosystem. In doing so, the book demonstrates what was lost when the lake was drained, and some of what might be reclaimed. 

Before it was drained, the lake provided critical habitat for a huge and diverse array of animals and plants. As Reimer explained, and the UBC study reiterated, the lake was home to large populations of salmon and sturgeon, while its surface and surrounding wetlands were home to huge populations of ducks, swans, blue herons, and geese. Restoring Sumas Lake could facilitate the return of those animals to the region, and provide important breeding and feeding grounds in ways that could help stimulate populations far beyond Abbotsford.

The lake would also facilitate the return of native grasses and trees, and stimulate an untold number of ecological processes and relationships that were compromised a century ago.

Environmentalists have detected a wealth of harmful substances in fish currently living in the waterways on Sumas Prairie. Those include agricultural products like pesticides, as well as human contaminants, including cocaine and painkillers. Restoring Sumas Lake would inevitably reduce the amount of fertilizer and other agricultural byproducts in local waters. And even if it wouldn’t eliminate human byproducts altogether, by dramatically increasing the amount of water on the landscape, the concentration of contaminants would be significantly reduced. 

Wetlands also help purify water by removing sediments and pollutants. A massive increase to the natural wetland environment in the Fraser Valley would help improve water quality locally and across the region.

Sumas Lake could also become a cornerstone in efforts to save animals at risk of local extinction. The UBC study notes that of 102 species at risk of extinction across the region, about one-quarter likely lived in the Sumas Lake area. 

📷 The Reach Gallery Archives P5659

People and reconciliation

The draining of Sumas Lake a century ago occurred over the objections of local Indigenous communities, with little consideration for the lake’s role in the lives and economies of the region’s long-time inhabitants, and with no compensation for the consequences of its elimination.

For millennia, the lives of locals had been closely tied to the lake. The Semá:th people built temporary homes on the lake, drawing on its bounty for food and natural resources. People ate fish from its waters, hunted the ducks that darkened the skies above it, and used the innards of sturgeon to create an early version of superglue. The lake’s waters provided a quick canoe route for locals to visit friends, families, and trading partners in other Stó:lō villages or to access fishing grounds in the Fraser River. 

Then the lake was gone. Its restoration wouldn’t right all historical wrongs, but it could be a pivotal piece in acknowledging  the value of what was lost. The UBC study suggests “Lake Back” would echo “land back” efforts elsewhere to restore land rights to Indigenous communities, while also bringing various positive environmental benefits.

The return of the lake could potentially foster salmon populations that would allow communities to undertake key ceremonial and cultural events and practices that have been threatened in recent years by low salmon returns and high water temperatures. 

Finally, the return of Sumas Lake could also have an impact on the final outcome of a decade-old land claim filed by Semá:th First Nation, seeking compensation for the draining of its former lake. The matter is still making its way through Canada’s tangled and much-delayed claims process, but Canadian governments seem likely to eventually be on the hook, one way or another, for the decision to drain the lake.

“The lake was … our fridge,” Semá:th Chief Dalton Silver said in 2013. “It meant everything to our people. They (the B.C. government) took away our culture, and lo and behold they took away our lake.”

Economic and financial

Environmentalists have long argued that there are real, tangible economic benefits that come from working with the natural environment, rather than against it. In Sumas Lake’s case, the most-obvious, non-hypothetical economic benefits would come from its ability to produce and foster greater fish populations.

Sumas Lake was both an important transit route for all species of Pacific salmon, and an important home for young fry. One study suggested between 230,000 to 23 million juvenile salmon could have reared in the lake. (That researcher noted that making a more exact estimate was nearly impossible.) Indigenous groups said a distinct population of Sumas Sockeye went extinct when the lake was drained. 

Sumas Lake represents around 7% of the total amount of floodplain habitat salmon have lost in the Lower Fraser watershed. That habitat provided important food, shelter, and rest for young salmon. Restoring the lake could help counter—whether in a small or large way—ongoing challenges to commercial fisheries both locally and in the ocean. Similarly, a reborn lake seems likely to help local sturgeon populations, which are relied upon by the numerous tourism-oriented sturgeon-fishing operations in the region.

Other economic and financial benefits related to an improved environment may be harder to measure, but could still have an important effect. The return of a cool lake could mitigate the region’s collective heat island impact, and increased greenery has been tied to better health outcomes. Those positive effects and others could provide both intrinsic and financial benefits. 

Most obviously, moving people and property off the lakebed would reduce the amount of property dependent on imperfect dikes to maintain businesses and homes. In doing so, it would remove the need for Canadian governments—and their taxpayers—to foot the bill for assisting lakebed-dwellers in the case of a future disaster.


The restoration of Sumas Lake would bring a handful of clear economic, social, and cultural benefits, even if their true scale and impact can’t quite be calculated.

But it would also facilitate a wealth of opportunities that are more speculative in nature. Their existence would depend on the initiative of individuals, organizations, and governments; their feasibility would be impacted by local, regional, and national politics, the state of global and more-local economies, and on environmental factors that remain uncertain. 

But the opportunities presented by a restored lake are large enough to be worth exploring and discussing, even if they are still conjecture.

Tourism and recreation

A restored Sumas Lake would bring significant, but likely highly variable, tourism potential.

Although lakes can be big business—just look at nearby Cultus Lake—a “natural” Sumas Lake restored along the lines of that envisioned by environmentalists would provide very different opportunities. Powerboat-oriented tourism is likely out of the question, but there would exist the potential for commercial endeavours—potentially driven by Semá:th First Nation—that would emphasize the lake’s beauty, history, and cultural importance. 

Wetland-oriented regional parks are popular and numerous throughout the Fraser Valley. Sumas Lake could present the opportunity to create a park that would be a regional, or even international, attraction. It could showcase the lake and surrounding valley, while teaching visitors about the past and future of the valley. 

On the northwest edge of the water’s typical boundary, the lake could form the springboard for commercial development on Semá:th First Nation’s reserve lands. At the lake’s southeastern edge, the farming community of Yarrow could also find the arrival of a lake bringing more visitors and tourists to their shops and businesses.

This potential, of course, would all be highly dependent on the configuration of the lake, how it was restored, the involvement of local First Nations, and land-use rules on its floodplain. (And on mosquitoes.)

There are other creative opportunities as well. Before the lake was drained, Semá:th people would occupy raised platforms in the middle of the lake during the summer to escape mosquitoes.. The return of structures to the lake could provide economic and cultural opportunities and facilitate significant tourism opportunities. (They could also create environmental risks seen as contrary to the lake’s restoration.)

📷 The Reach Gallery Archives/P188

Cultural and historical

Whether or not tourism is the focus, the lake would bring massive cultural and historical opportunities to the Fraser Valley. Its restoration would provide a visual centrepiece for potential museums, cultural centres, and interpretive attractions. It could also serve as the basis for larger cultural and historical development in the region. 

Sumas Lake was not a lake stuck in a prehistoric and pre-human past—people had lived near Sumas Lake for thousands of years. Artifacts from an important village at the junction of the Sumas and Fraser rivers show the Semá:th people were likely living there around the same time that Alexander the Great was leading Greece, and continued living there until 1906. Reimer writes in Before We Lost The Lake that, in the 1840s, a surveyor found a local Indigenous community had built a “large stockade fort” that resembled those built elsewhere in the country by the Hudson Bay Company. 

The return of the lake could facilitate the re-emergence and reconstruction of pieces of the valley’s history that few people today know exist.

Semá:th Chief Dalton Silver 📷 Tyler Olsen


The return of Sumas Lake will require buy-in and agreements between a dizzying array of partners. That’s a major challenge. But it could provide an opportunity.

If the lake can be restored in a way that is acceptable to residents in the area, its return could provide a template for how future agreements elsewhere could work and show how efforts at reconciliation could provide win-win benefits. 

But Semá:th Chief Dalton Silver has urged caution and the need to respect current landowners. He has spoken at length about the damage caused by the draining of the lake, but also has repeatedly said that he does not want to force property owners off their land to accommodate its return. 

“I worked with BC First Nations on land claims and my grandfather did as well,” Silver told The Current in 2022. “We don’t want to do it in such a way that would force people from their homes. My grandfather said, ‘We won’t do that because we know how that feels.’”


or to participate.