The season so far · Tuesday, September 29, 2026
The lowest water in thirty years
The Rhine gauge at Kaub read 5 centimeters this morning. Since January the Danube has spent 157 days below the flow where ships start to have trouble, and the Rhine 140. Our records go back to 1997, and no year in them comes close to this one.
If you have a cruise booked, that sounds like bad news. Usually it isn't, and it's worth a minute to say why.
How this year compares
We count a low-water day the same way in every year. When the river drops below the flow where ships start having trouble at its shallowest point, that day counts. The rule for 1998 is the rule for 2026. The counts below run to September 29 in every year, so a season that isn't finished is compared with the same stretch of the ones before it.
Rhine
highest on record
140
19972026
140 days so far. The old high was 92 in 2003.
Danube
highest on record
157
19972026
157 days so far. The old high was 76 in 1998.
Elbe
highest on record
74
19972026
74 days so far. The old high was 57 in 2003.
The Danube is the one that stands out. Its worst year before this was 1998, with 76 days by September 29. This year it has 157. The Rhine is 48 days past 2003, and the Elbe is 17 days past 2003.
The Seine is having an ordinary year. Its trouble is high water rather than low, and it sits 11th of 30, which is about the middle of the pack.
What it means if you're sailing
Low water rarely cancels a cruise. Most of the time the ship carries less weight through the shallow stretch and you notice nothing at all. If the water keeps dropping, the line moves you onto a sister ship waiting on the other side of the shallow part, with a coach ride in between. In a longer spell the ship stays where it is and you see the towns by coach. Cancellations do happen, but they are rare.
The lines have had practice. The Rhine had hard years in 2003 and 2022, and the ship swaps that once felt improvised are now something crews do as a matter of routine. If you want the detail, we wrote it all out in what low water means.
Has the Rhine been this low before?
Yes. The Lower Rhine at Lobith sat at almost exactly today's flow in 2018, within 3.8% of where it is now.
Here is what the river did over the six weeks after each of the closest years. Nothing here is a forecast. It's the measured record.
| Year | Date | Flow | vs today | Back over the line | Days low of the next 42 |
|---|---|---|---|---|---|
| 2018 | October 19 | 642 m³/s | +3.8% | not in six weeks | 42 |
| 2003 | September 27 | 737 m³/s | +19.2% | after 11 days | 24 |
| 2009 | October 5 | 797 m³/s | +28.9% | after 8 days | 16 |
| 2016 | October 18 | 835 m³/s | +35% | after 5 days | 10 |
| 1997 | October 7 | 854 m³/s | +38.1% | after 5 days | 17 |
The pattern is the useful part. In every one of those years the river climbed back over the line, and it took between 5 and 11 days to do it. What it didn't do was stay there: in the harder years it dropped back under repeatedly through the autumn. That is why a cruise in a low-water spell usually means a rerouted day or two rather than a lost holiday.
Flow figures are the European Union's GloFAS record for Lower Rhine at Lobith, measured on September 28 and the same date in each earlier year. The line is the level below which this stretch has run only 12% of the time in thirty years.
When the water comes back
A wet week helps a little. What actually refills these rivers is winter rain, and then the Alpine snowmelt that runs down through May and June. That pattern hasn't changed. You can see it in the Rhine's next twelve months below: the odds ease through the winter and the spring months look ordinary again.
Sep*
Oct*
Nov*
Dec*
Jan*
Feb*
Mar*
Apr
May
Jun
Jul
Aug
If you are sailing next year, this summer doesn't tell you much about your dates.
If you're sailing soon
Check your dates. You get the odds for your exact window, day by day, and what the same dates did in each of the last five years. If your cruise is further out, you can leave us your email on that page and we'll write to you if the outlook for your dates changes.
Where these numbers come from
The counts on this page come from the European Union's GloFAS river flow record, which runs from 1997 to today, and they are worked out again every morning. This morning's gauge readings come from WSV PEGELONLINE in Germany and Vigicrues in France. The method is written up in full on how it works.
One thing worth saying plainly. These counts are modeled river flow, not the numbers painted on the gauge boards at Kaub or Dresden. Flow is the fairer way to put thirty years on one scale, but it does mean our count for an older year can differ from what a particular gauge showed in a particular week.
Journalists and researchers are welcome to the underlying figures: departures.json, rivers.json and gauges.json are open, and this page explains how to cite them. Page last updated September 29, 2026.