Why Nijmegen built an island to give floodwater another route

Nijmegen · Living with rivers

The city made room for water by changing the land

Nijmegen’s river island is part of a flood-management landscape. Its value becomes clearer when you look beyond the waterfront photograph.

1. The extra route is the interventionMoving the dike inland created space for a bypass beside the Waal’s narrow bend.
2. Some park routes are meant to get wetPublic access and flood conveyance do not have to occupy the same space at every water level.
3. A natural-looking channel still needs careSediment management belongs to the operating story, alongside the original landscape design.

At Nijmegen, a green island sits between the city and a second ribbon of river. From a distance, it looks like an ambitious waterfront improvement: bridges, banks, paths and somewhere to spend an afternoon. The surprising part is that creating this island was also a way of giving floodwater more room.

The project is often attractive enough to explain itself badly. A photograph suggests that the Dutch solved a water problem by adding a park. The more useful story is about changing where water can pass, which pieces of land can temporarily become wet, and what must happen when the river starts rearranging the new channel.

That last part matters. The maintenance record is not a footnote to the beautiful design. Read beside the original plans, it helps distinguish an adaptable landscape from the fantasy of flood protection that looks after itself.

The island is what remained when the river gained a second route

The Waal bends sharply and narrows near Nijmegen. According to H+N+S, the project’s landscape architect, moving the dike at Lent 350 metres inland made room for a bypass carrying water during very high river levels. An elongated inhabited island emerged between that channel and the main river.

The island therefore makes more sense as the land between two routes than as an object dropped into the river. What changed was the surrounding geometry. Water gained a way around the constrained reach, while the city gained a new relationship with land that was no longer simply behind the old dike.

Rijkswaterstaat’s description of river measures uses Lent as an example of moving a dike and adding a channel to improve high-water flow. It places that intervention among several tools, including removing obstructions and improving dikes where widening is impractical.

More room can mean passage rather than a bigger bucket

“Room for water” can describe two different services. Both matter, but confusing them makes it difficult to understand what a particular project is supposed to do.

Rijkswaterstaat explicitly treats temporary storage as a separate measure, giving the Volkerak-Zoommeer as an example when storm-surge barriers obstruct discharge to sea. Nijmegen’s bypass should not be imagined only as a container that fills up and then stops helping. The connected flow path is central.

This is a useful distinction when comparing infrastructure stories. In Pornainen’s thermal store, capacity divided by demand explains an idealised duration. That same “how long until the container is empty?” question would miss much of what a river bypass is for. The physical service must determine the question.

A submerged path can be part of the intended landscape

H+N+S describes paths at different elevations, an inlet threshold containing tubes for water renewal, and a Citadel bridge access route designed to be floodable. These are deliberate features, not evidence that the designers forgot the water would rise.

A wet path and a failed flood defence are different observations. To interpret the first, you need to know what that piece of land was designed to do.

The practical consequence is a changing map of use. A space available to people at one water level can serve the river at another. Its temporary loss as a walking route is not automatically a loss of its underlying function.

That does not mean visitors should treat rising water as harmless or cross submerged access routes. Design intent is not a live safety assessment. Closures and official instructions still govern access; a landscape photograph cannot tell you whether a particular route is safe today.

Nor does a floodable park certify the safety of every surrounding property. The inference is narrower: intentional inundation exists, so water on a path is insufficient evidence of a failed protective system. You need the site’s design and the actual event conditions before drawing that conclusion.

The maintenance record is more revealing than the postcard

Team Rijntakken’s 2024–25 maintenance record says the Spiegelwaal still met its design, while identifying sediment accumulation east of the inlets. The Waal had carried that material in.

Excavation equipment changed as falling water exposed the working area. The schedule envisaged removing about 25,000 cubic metres by early 2025—a planned quantity, not an independently verified completion total.

The distinction between the affected inlet-side reach and the whole Spiegelwaal is essential. It would be misleading to turn this record into “the Dutch river park clogged up and failed.” It would be equally misleading to omit the intervention and present the channel as maintenance-free.

The river carries more than water. A design that makes room for flow must also confront material arriving, settling and changing the available space. Here, that issue appears in a concrete work plan, complete with equipment choices that respond to the same fluctuating water levels the park was designed around.

The useful test for a nature-based infrastructure claim is therefore not “Does anybody ever have to intervene?” Ask whether the required intervention is understood, assigned and compatible with the purpose of the project. Maintenance can be evidence of stewardship; whether it is adequate requires more than the existence of a web page.

Giving the river space is also deciding what people can do with land

The municipality’s 2012 official notice describes the relocated dike, excavated channel and island, alongside retained buildings and proposed uses. It also lists the planning, water and environmental decisions required to enable the project.

That record changes the familiar slogan. Space was not simply discovered beside the river. Public authorities had to decide what the land would become and which activities could coexist there. A technical drawing could show the desired channel, but it could not by itself settle all the uses of the surrounding ground.

This is also why the project cannot be exported as “every city should move its dikes back.” Availability of land, existing settlement, river behaviour and the rest of the protection system all matter. Rijkswaterstaat’s own toolkit retains dike strengthening where widening is not feasible.

There is a related distinction in Poland’s renewable-energy planning maps: identifying technical potential does not automatically resolve land-use and environmental decisions. The domains differ, but the gap between an attractive spatial idea and an authorised, workable project is real in both.

And changing a hazard’s physical route is different from observing it. Our article on a submarine cable sensing a tsunami concerns information about a moving disturbance. Nijmegen concerns how a particular landscape accommodates high water. Neither function should be mistaken for a guarantee that the hazard has disappeared.

The island’s beauty is not incidental: usable public space helps make a major river intervention part of a city’s daily life. But its most instructive feature is the relationship between the ordinary day and the exceptional one. The same landscape can welcome people, give water passage and require a maintenance crew—without any of those roles cancelling the others.

Questions about Nijmegen’s river island

Project records and official sources

Primary project and maintenance records checked 6 September 2026

• H+N+S — Room for the River Nijmegen — Landscape design, setback, threshold, access and visual references.
• Rijkswaterstaat — Maatregelen rivierengebied — Distinct river-management interventions.
• Team Rijntakken — Veur-Lent maintenance — Local sediment problem, equipment and stated work schedule.
• Staatscourant 2012, 6680 — Official Nijmegen planning and project notice.
• Design descriptions are not independent measurements of present flood risk. The maintenance schedule is not treated as a verified completion certificate; no fixed water-level reduction is claimed.