Floods are Natural. Our Disasters are not.
- Andreas Kondos Sheppard
- Dec 13, 2025
- 11 min read
Andreas Kondos Sheppard
Imagine a river mid-flood.
Not the catastrophic images of our modern era; just imagine what a natural river honestly looks like mid-flood.
The vast, churning water, rich in brown, silty sediment. The embankments as they erode. The sound of the flow and its power, pouring over to the shallow marshes flanking its banks. Trees, knee-deep in water. Temporary ponds forming as water fills depressions in the earth.
Somewhere upriver, weeks, months, or even years’ worth of water has been unleashed. But these downstream sections do something that money cannot: buy the river time (Bullock and Acreman).
Time to address this volume of water. To hold it like a sponge and release it gradually, whilst forming crucial habitat in the process. Time for the flow to slow and the river to deposit its nutrient-rich sediments onto the surrounding land. For centuries, humanity has relied on these events to fertilize soil and breathe life into the environment (Zedler and Kercher).
We call these places wetlands and floodplains. The places that water has entered before and will enter again. In the way the shocks of a car absorb the impact of a bump, these places absorb the impact of the flood –– to the benefit of everything downstream –– by reducing the peak discharge and delaying the surge (Bullock and Acreman). These spaces, in a way, absorb the shock of a flood.
Now imagine we entomb these lands in cement.
We drain the wetlands and floodplains; straighten the river’s course; reinforce the embankment with concrete; add levees to avoid waters breaching into surrounding land; and we call these improvements. We “reclaim” land from the river, building entire communities atop these ancient river shocks (Burby). And over time, the people forget. They forget what the land once was, and the remarkable effort needed to clear it. They forget the natural rhythm of the river. And they forget exactly what the land they now live on was doing in the first place.
But the river does not care what we forget. When the excess water upstream surges downwards, there is no land prepared to soak up the excess like a mop. Rather, the straight, narrow channels we build for the river to flow through become overwhelmed, and the water has nowhere to go but out. And it does go out: into surrounding communities, communities built atop the river’s shocks. And without any buffering, the full force of the surge is felt, destroying homes, infrastructure, and entire communities (Burby; Canadian Climate Institute, High and Dry). It’s as though the river itself is trying to destroy those things built in places they should not have been.
And this pattern shows up, almost word-for-word, in our modern floods around the world (Canadian Climate Institute, Turning the Tide).
The Lower Fraser, 1948.
At this point in history, the broad floodplain of the Fraser River had become a quilt of farms, towns, and industrial centres, sewn together with roads and railways. But upriver, the water was stirring. A particularly large snowpack had developed that winter deep within interior B.C., and it lasted far longer than usual. When the spring cold snap ended, the snowpack melted unusually fast, and was bolstered by large spring showers. This water surge saw the river rupture dikes from Agassiz to Matsqui, flooding roughly 50,000 acres of the Lower Fraser Valley. About 16,000 people were forced from their homes, and later reports would suggest that much of the damage was from “intensive development” on the floodplain itself (“From the Archives: The 1894 and 1948 Fraser Valley Floods.”)
Vanport, 1948
That same year, another similar story unfolded –– this time on the mighty Columbia. Vanport, Oregon, built on a low-lying pocket of drained wetlands and a former lakebed sandwiched between the Columbia and Willamette Rivers, would bear the brunt of this flood. Protected only by earthen levees and railroad embankments, flooding was a matter of when, not if. And on May 30th, 1948, after weeks of high water, an embankment failed. In mere hours, Vanport was destroyed, and 18,500 people –– a third of them African Americans –– were left homeless. Many of these families worked in shipyards, which themselves existed on land “reclaimed” from the Columbia (McGregor; Ott).
The Mississippi, 1993.
Perhaps the greatest example of poorly designed flood infrastructure is on the enormous Mississippi River system. Throughout the 20th century, engineers worked to straighten, levee, and entrap the river into increasingly small areas to make way for towns, farms, and industry –– much of this constructed on former floodplains. In 1993, the Great Flood of the Mississippi River occurred. Over 1,000 levees across the basin failed or overtopped, with enormous damage occurring outside protected areas; the water was pushed higher and faster between the levee walls (U.S. Geological Survey; National Weather Service). The levees had funnelled the river’s fury to communities without levees. Since then, it has been shown that structures at the margins of leveed floodplains can actually be worse off than if no levees had existed at all (Pinter et al.). The illusion of safety presented by the levees led to a false sense of security for inhabitants; and tragedy followed.
And climate change is not helping.
The physics are simple: warmer air holds moisture better, which directly leads to increased precipitation (U.S. Environmental Protection Agency, Climate Change Indicators: Heavy Precipitation). This precipitation worsens flooding events and increases river discharge, meaning flood control infrastructure requires continuous care and repair. Unfortunately, this is not being done. Many of the developments along river systems are designed to hold water discharge levels from 50 or more years ago, and are falling apart due to age (Milly et al.; Canadian Institute for Climate Choices, Under Water). Floods that were once “once-in-a-lifetime” are becoming “once-in-a-decade” (Canadian Climate Institute, “Fact Sheet: Floods”; U.S. Environmental Protection Agency, Climate Change Indicators: River Flooding). These systems are not equipped to deal with the volume of water modern rivers can discharge in floods. Add to this that climate change can cause unpredictable rain and drought patterns, and predicting the severity of incoming floods becomes a Herculean task (Intergovernmental Panel on Climate Change). And this doesn’t even account for the additional damage other climate-change-impacted weather patterns can have on flood infrastructure (i.e., hurricanes, tornadoes, etc.).
All of this does not compound to “a little more flooding.” No. This compounds to catastrophic flooding events, where streets act as firehoses for floodwater, blasting into neighbourhoods living under the false veil of safety because of a levee marked on a colourful brochure map (Burby). This false sense of security is deadly; the very existence of these communities is itself a flood hazard (Burby; Canadian Climate Institute, High and Dry).
If this were a story of doom-and-gloom, it would end here. The climate warmed, the floods worsened, and people died.
But it’s not. This story in particular is one caused by humans, and it can be fixed by humans too.
For decades, we have viewed wetlands and floodplains as things to be used, developed, and “improved.” Local governments saw in wetlands and floodplains usable land; land that could be populated by tax-paying constituents and profitable industry. They did not see the role that the land itself was playing as flood-control infrastructure (Bullock and Acreman; Zedler and Kercher). Of the floodplain maps that exist (which are remarkably few), many are outdated and do not reflect the degradation of wetlands surrounding them, the increase in flood discharge, and the people who now live in these zones (Canadian Climate Institute, Turning the Tide). Many of the constituents who live in these zones don’t even know the risk (Canadian Climate Institute, Turning the Tide). And updating these maps today runs the risk of forcing uncomfortable conversations: about usable land, restoration, and risks the people who now live on these zones face. Such conversations are the politician’s poison: making people uncomfortable is not a great reelection strategy (Canadian Climate Institute, Turning the Tide).
On paper, governments have taken action to address this. Goals like “no net loss of wetlands” were set out, subsidized flood insurance was given, and generous disaster relief funds set up. But in practice, these policies have often worsened the issue. Destroying wetland in one place and recreating it in another does not recognize what the first wetland was doing: acting as flood control for the region. Often, new wetlands are put atop other ecosystems, destroying one ecosystem to save another. And even then, these new wetlands ecologically underperform –– the wrong environments in the wrong places (Turner, Redmond, and Zedler; Zedler and Kercher).
Subsidized flood insurance and disaster relief funds ease concerns over flood impact, and paradoxically entice people to move into high-risk zones. If a flood does come, there is a “safety net” in place; but these safety nets often fail. The floods are worse than expected, and end up costing taxpayers absurd quantities of avoidable spending (Burby; Canadian Climate Institute, High and Dry).
So… what does doing better look like then?
We can turn to the Netherlands as an example. In the 1990s, several near-miss floods on the Rhine and Meuse rivers forced Dutch planners to reconsider their approach of constantly raising the levees. Instead, they pioneered the Room for the River program (Rijke et al.).
Over 30 interventions to lower floodplain surfaces occurred; side channels were dug; old overflow paths reconnected; and dikes moved or set back. By adding approximately 4,400 hectares to floodplains on the Rhine and its branch channels, they allowed floods that would have been devastating to slowly and manageably flow to the sea (Klijn, Asselman, and Wagenaar).
Other initiatives are sometimes even simpler: removing levees to allow rivers to access side channels; purchasing homes that flood every few years and converting the land back to its natural form. Often, moving people to new regions and allowing the ecosystem to recover proves more economically sensible than to fight a losing battle against water, the climate, and time (Mach et al.).
Using an example closer to home, take the Sumas Prairie in the Fraser Valley. Before the prairie, there was a lake: Sumas Lake (Xwōxtsa). Until the 1920s, this lake was a dynamic floodplain that expanded and shrank with the seasonal waters, supporting enormous salmon runs, waterfowl, and wetland ecosystems (Finn et al.). However, not recognizing its importance, B.C. engineers drained it for “usable” farmland, and interspersed the landscape with canals, levees, and river diversions to keep the lakebed dry. About a century later, in November of 2021, an atmospheric river reminded the landscape of what was once lost. Dikes burst and water flowed, and for a few days, the “reclaimed” lake returned, in defiance of human engineering, whilst in the process displacing the region’s residents (Finn et al.).
Afterwards, Indigenous groups, non-Indigenous researchers, land stewards, and legal analysts proposed a radical idea: bringing the lake back. Evidence suggested that restoring the lake through property buyouts and reclamation efforts would be approximately equivalent in cost to upgrading existing levees and pumps (Finn et al.; “Researchers Call for Return of Sumas Lake”). Although the land has not yet been returned to its original state, this case demonstrates how mindsets have begun to shift towards working with the land, and those who have traditionally stewarded it, to manage flooding (Finn et al.).
The idea of working with the water and the land –– rather than against it –– is what smart flood policy looks like (Rijke et al.; Mach et al.).
We do not need to tear down every levee and move every community. This is simply not feasible. Carefully placed storm sewers and flood walls can absolutely be effective for controlling local floods, especially in developed places that naturally receive a lot of rain. The water has to drain somewhere. Ports built atop coastal wetlands and floodplains are economically vital and generally impossible to move. But modern hydrological research shows that we can best manage floods when we integrate our engineered solutions with those nature already has –– to work with the land rather than against it (Bullock and Acreman; Milly et al.; Rijke et al.).
In practice, this might look like protecting remaining wetlands and restoring crucial ones (Zedler and Kercher). Using money set aside for flood relief and insurance to instead help move people into safer regions and restore the landscape –– fixing the problem at its root, rather than slapping a band-aid on it (Mach et al.; Canadian Climate Institute, High and Dry). Updating maps of remaining floodplains and wetlands, while providing more accurate maps of flood-risk zones (Canadian Climate Institute, Turning the Tide). Saying “No!”, however difficult it may be, to developments on lands that will inevitably flood (Canadian Climate Institute, Turning the Tide).
We built atop the lands that silently did the work for flood control, and gave the flood waters nowhere else to go. In looking for new paths, the waters found us. Again, and again, and again (Bullock and Acreman; Burby).
We cannot control the weather, storms, or the river. We can try to, but nature inevitably will win. Climate change has loaded the dice for heavier rainfall, more intense flood surges, and more record-breaking headlines. That is inevitable at this point (Intergovernmental Panel on Climate Change; Milly et al.). But we do control where we build, what we build, what we protect, and whether we work with or against the land in controlling future floods (Canadian Climate Institute, Turning the Tide; Rijke et al.; Mach et al.).
Floods are an inevitable part of river systems. The damage they do to us is not (Burby; Canadian Climate Institute, High and Dry).
Works Cited
Bullock, Andrew, and Mike Acreman. “The Role of Wetlands in the Hydrological Cycle.”
Hydrology and Earth System Sciences, vol. 7, no. 3, 2003, pp. 358–389,
Burby, Raymond J. “Hurricane Katrina and the Paradoxes of Government Disaster Policy: Bringing About Wise Governmental Decisions for Hazardous Areas.” The Annals of the American Academy of Political and Social Science, vol. 604, no. 1, 2006, pp. 171–191.
Canadian Climate Institute. “Fact Sheet: Floods.” Climate Change and Floods, Sept. 2024,
Canadian Climate Institute. High and Dry: The Rising Tide of Flood Risks and the Insurance Dilemma. By Ryan Ness and Camila Florez Bossio, 18 Mar. 2024,
Canadian Institute for Climate Choices. Under Water: The Costs of Climate Change for
Canada’s Infrastructure. 2020.
Canadian Climate Institute. Turning the Tide: How Flood Risk Transparency Can Drive
Equitable Outcomes in Canada. By Erik Sparling, Jimena Eyzaguirre, and Janna Wale,
Canadian Climate Institute, Jan. 2024,
Finn, Riley J. R., et al. “Reclaiming the Xhotsa: Climate Adaptation and Ecosystem
Restoration via the Return of Sumas Lake.” Frontiers in Conservation Science, vol. 5, 2024, article 1380083,
Luymes, Glenda. “From the Archives: The 1894 and 1948 Fraser Valley Floods.” Vancouver Sun, 17 Nov. 2021,
Intergovernmental Panel on Climate Change. Climate Change 2021: The Physical Science Basis. Cambridge UP, 2021.
Klijn, Frans, Nathalie Asselman, and Dennis Wagenaar. “Room for Rivers: Risk Reduction by Enhancing the Flood Conveyance Capacity of The Netherlands’ Large Rivers.”Geosciences, vol. 8, no. 6, 2018, article 224, doi:10.3390/geosciences8060224.
McGregor, Michael. “Vanport.” The Oregon Encyclopedia, Oregon Historical Society,
Mach, Katharine J., et al. “Managed Retreat through Voluntary Buyouts of Flood-Prone
Properties.” Science Advances, vol. 5, no. 10, 2019, article eaax8995, doi:10.1126/sciadv.aax8995.
Milly, Paul C. D., et al. “Stationarity Is Dead: Whither Water Management?” Science, vol.319, 2008, pp. 573–574.
National Weather Service. “The Great USA Flood of 1993.” Northwest River Forecast
Ott, John. “Vanport Flood Begins on Columbia River on May 30, 1948.”HistoryLink.org,
Pinter, Nicholas, et al. “Modeling Residual Flood Risk Behind Levees, Upper Mississippi
River, USA.” Environmental Science & Policy, vol. 58, 2016, pp. 131–140.
“Researchers Call for Return of Sumas Lake Following Devastating 2021 Floods.” UBC
Faculty of Forestry, University of British Columbia,
Rijke, Jessica, et al. “Room for the River: Delivering Integrated River Basin Management in the Netherlands.” International Journal of River Basin Management, vol. 10, no. 4, 2012, pp. 369–382.
Turner, R. Eugene, Armand M. Redmond, and Joy B. Zedler. “Count It by Acre or
Function—Mitigation Adds Up to Net Loss of Wetlands.” National Wetlands Newsletter, vol. 23, no. 6, 2001, pp. 5–6, 11–12.
U.S. Environmental Protection Agency. “Climate Change Indicators: Heavy Precipitation.” Climate Change Indicators in the United States, 2021,
—. “Climate Change Indicators: River Flooding.” Climate Change Indicators in the United
States, 2021,
U.S. Geological Survey. The Great Flood of 1993 on the Upper Mississippi River—10 Years Later. Fact Sheet 2004–3024, 2004, https://pubs.usgs.gov/fs/2004/3024/report.pdf.
Zedler, Joy B., and Suzanne Kercher. “Wetland Resources: Status, Trends, Ecosystem
Services, and Restorability.” Annual Review of Environment and Resources, vol. 30, 2005, pp. 39–74
Comments