This is the public record of the 2026 low water on the Rhine and Danube — the numbers, what they cost, and the signal chain that was already visible in April 2025. Every figure below is sourced.
The same signal — record low water — lands as a different number in every sector. These are the 2026 figures as reported.
| Who | What happened | Figure | Source |
|---|---|---|---|
| Inland shipping | Barge freight Rotterdam → Karlsruhe, June to August | €45 → €150–160 / tonne | Breakbulk News; CLECAT (Reuters) |
| Barge loading | Load per vessel at Kaub, versus 1,200 t downstream | ~220 t | CLECAT; Breakbulk News |
| German industry | Thirty days of low water, modelled effect | −25% inland volume · −1% output | Kiel Institute |
| German GDP | Q3 drag if low water persisted | −0.1 to −0.2 pp | Kiel Institute (Kooths) |
| Chemicals & steel | thyssenkrupp Steel halted its own barging; BASF declared force majeure on surfactants | production cut | transportchronicle.com; TrasportoEuropa |
| Nuclear | Romania Cernavodă 1 & 2 down; Hungary Paks 3 of 4 units off | 1.4 GW · 500 MW left | Electricity.Trade; Ember |
| French nuclear | Capacity lost to environmental constraints in July | −18% | Ember |
| Manufacturing | Dacia and Ford halted production in Romania to cut load | halted | EnergyReader |
| French farms | National crop and livestock losses, 2026 | > €10 bn | FNSEA via Le Monde; Politico |
| French vineyards | Harvest versus the five-year average | −17% · lowest since 1957 | OIV via SBS News; Vinetur |
| River cruises | Reroutes, ship swaps, 1,500 km coach transfers, cancelled sailings | sailings lost | Travel Extra; Cruises News |
| Riverfront towns | Day spend lost where ships skipped a call | income down | The Traveler |
Figures are 2026 public reporting values; hydrological readings should be taken from the PEGELONLINE / WSV gauge records. This page is a directional reading of public data, not investment advice.
None of this was unforeseen. Here is the chain, in public data, in order.
Not a failure to predict. A failure to respond. The signal, the expert warning and the official probability forecast all arrived — and the network still did not move in time. Rail was under works until December; road hauliers reported no trucks to be found.
A directional read has to keep two things apart: the direction a signal pushes demand, and whether a given operator is ready to capture it.
| Layer | What 2026 showed | Reading |
|---|---|---|
| Direction | With barges cut back, cargo moved en masse to road and rail — the only alternative, and the rates on those routes rose with it. | real, and large |
| Capture | Rail was closed for works until December; hauliers could not find trucks. One barge's load needs 100–200 trucks. | not ready |
The gap is not in the signal. It is in the time between seeing the signal and being able to act on it. That gap is what a lead time is for.
These losses were avoidable. The signal was always there — not hidden, just never connected to you.
We do not claim to predict better than the official services. We take the signal that already exists — snow, ice, soil moisture, discharge — and put it in front of the people who set capacity, routes and price, while there is still time to move.