Four months of block driving under the Western Scheldt: accessibility becomes a rationing question

Under the Western Scheldt, in the eastern tube of the tunnel between Terneuzen and Ellewoutsdijk, a crack of roughly thirty metres runs across fifteen tunnel rings. The only way to repair it is to empty the tube and replace concrete for four months. That work starts on 11 January 2027. What begins is less a construction project than a rationing problem: how many vehicles can still pass through the remaining tube, and who decides which ones.

Since last week that question has become concrete. On 31 August the permit desk opened for organisations, with a first deadline of 20 September. And earlier this month the Provincial Council of Zeeland voted unanimously to double the accessibility budget, from roughly 11 million to 22.1 million euros. Both decisions are about the same thing: what it costs to keep a region reachable when its main connection drops out.

A tunnel with no alternative

The Western Scheldt Tunnel is 6.6 kilometres long and carries around 30,000 vehicles on an average day. Since 30 December 2024 it has been toll free for passenger traffic, which has pushed usage up further. For Zeeuws-Vlaanderen it is not one connection among several but the connection: the only fixed link to the rest of Zeeland and therefore to the rest of the Netherlands. Anyone who cannot get through either detours via Antwerp or takes the passenger ferry between Vlissingen and Breskens, which does not carry cars.

The concrete damage in the eastern tube was discovered two years ago. The repair was subsequently postponed, partly to work out the traffic approach in more detail. In hindsight that delay was sound, because the first solution on the table turned out to be untenable.

Why the detour was dropped

In January of this year two scenarios were under consideration. In the first, the western tube would carry traffic only from north to south, with a short interruption each hour for convoys of emergency services and public transport in the opposite direction. All other traffic would have to detour via Antwerp. In the second, the western tube would alternate direction, with a reservation system for anyone wanting to use it.

What stands out is that the first scenario did not fall on Dutch objections but on Flemish ones. Employers’ organisation Voka calculated that the detour would put roughly 15,000 additional vehicles a day onto the Antwerp ring road, precisely during the Oosterweel works there. Its managing director Luc Luwel called the eventual decision to go with block driving the most appropriate and logical solution. Zeeland thereby opted for a variant that causes more disruption to its own residents, but does not push the problem across the border.

That is a lesson still missing from too many Dutch project organisations: a project’s environment does not stop at the national boundary, and a detour route is an intervention in the area it runs through. Design a detour without involving the road authority and the stakeholders on the other side, and you are designing a decision that will later be reversed anyway.

Block driving, and why two-way traffic is not an option

The chosen approach is known as block driving. For roughly fifty minutes traffic runs through the western tube in one direction, after which the tunnel closes for ten minutes to reverse all signalling, signage and lane closures. Then traffic runs the other way. Estimates put throughput at 700 to 800 cars per hour, and that number drops as soon as lorries are included.

The obvious question is why traffic cannot simply run in both directions through the same tube, as it does in Swiss mountain tunnels. The answer lies in ventilation. Tunnel director Harald Schoenmakers explained that the ventilation system moves air in the direction of travel. In a fire, two-way traffic means that some road users end up standing in dense smoke on the wrong side of the seat of the fire. For an immersed tunnel, where escape is only possible horizontally, that is unacceptable.

This is exactly the kind of argument that has to be explained carefully in communication with the public. A driver waiting four months for a time slot sees an empty carriageway beside them and draws their own conclusion. If the safety reasoning is not told in plain language and repeated, the measure reads as bureaucracy rather than as a safety requirement.

The permit system is the real stakeholder challenge

During the repair, available capacity drops by 50 to 80 per cent, with the upper end of that range applying to peak hours. So there is no version of this project in which everyone who travels today can still get through. The province has therefore built a permit system that sorts traffic into eight categories, numbered 0 to 7.

Category 0 covers the emergency services, which always have access. After that come urgent traffic, scheduled public transport, collective transport, freight, essential work traffic, compelling personal circumstances and, in last place, general private traffic.

Allocation runs in phases. In the first phase, which closes on 20 September, hauliers and essential work traffic can apply for a fixed recurring trip covering the whole works period. October brings a round for individual trips, with priority for collective transport and a cap on the number of trips per organisation per day. Where applications exceed capacity, a lottery decides. From 14 December the lottery is dropped and whatever remains is allocated first come, first served. Only from 21 December can private individuals apply, up to four hours before departure.

For stakeholder and environmental management professionals this is the most interesting part of the project, and the most underestimated. A categorised permit system is not a traffic measure but an allocation decision, with allocation politics attached. Every category boundary is a place where someone falls just outside it: the home care organisation that turns out not to qualify as essential work traffic, the contractor whose crew is not recognised, the pupil with no bus service. Anyone building such a system needs three things in place: clear criteria published in advance, an appeal route that genuinely works within days, and a safety net for the cases nobody anticipated. The third is almost always the weak spot.

There is a practical obstacle on top of that, one routinely overlooked in systems of this kind: applications run through eHerkenning, the Dutch business identity scheme, with a single email address per organisation that cannot be changed afterwards. For large employers that is a formality. For small businesses, care providers with dispersed teams and voluntary associations it is a threshold that quietly determines who gets through. In this project, digital accessibility has become a physical accessibility issue.

The mitigation package doubles, and the state looks on

The repair itself costs close to 20 million euros. The package to keep the region reachable in the meantime now costs 22.1 million euros, against the roughly 11 million originally assumed. It covers shuttle buses, temporary park and ride sites on both banks of the Western Scheldt, additional sailings of the Westerschelde Ferry, which will run every half hour during the works, and the reservation system itself.

The figure deserves attention, because the ratio is striking. On this project, keeping the surroundings accessible costs more than the physical intervention. That is no longer exceptional on renewal and renovation projects involving links without an alternative, and it should be treated for what it is: a full budget line that belongs in the estimate from the exploratory phase onward, not an item that appears during delivery.

Funding is not settled. Provincial executive member Harry van der Maas has repeatedly asked the Ministry of Infrastructure and Water Management for a contribution and has so far been told no. In the council session, the BBB, VVD and PvdA-GroenLinks groups called that unacceptable. Formally there is something to be said for the state’s position: the tunnel is provincial property and the province is its operator. Substantively it grates, because the chosen variant was partly driven by the interest in sparing the Antwerp ring road, and because since the removal of tolls the tunnel effectively functions as a national through route. Whoever helps decide the variant will not, in the long run, escape a conversation about the bill.

What this project shows

The Western Scheldt Tunnel is not an exception but a preview. Over the coming years the Dutch renewal and renovation programme will increasingly hit assets with no redundancy: one bridge, one tunnel, one lock, and no second route of comparable quality. In those places the centre of gravity of stakeholder and environmental management shifts from mitigating disruption to allocating capacity.

That calls for three things that are still far from standard. First, start the conversation about allocating residual capacity early, during the exploratory phase, not three months before the closure. Second, budget the mitigation package seriously and settle early who pays for it, because arguing about that in the year of delivery costs time that is not available. Third, involve road authorities and stakeholders at the far end of every detour route from the outset, including when they sit on the other side of a national border.

Zeeland has largely got the first and third right. The second is still open. That the funding remains unresolved four months before the closure is the one real risk in an otherwise carefully constructed plan.