In Vang Vieng We Got Around by Boat. Days Later the Same Water Was in Vientiane. How Upstream Flooding Actually Works in Laos.
There were days working as a guide in Laos when the answer was simply no. Not a modified itinerary, not an indoor alternative. The rain was heavy enough that nobody was going anywhere, and the only honest thing to tell people was that today was not happening.
Then there was August 2023, when the water did not stop at heavy rain. Vang Vieng went under. Not a street or two near the river, the town. People moved around by boat, down roads that had been roads the day before. I was living in Vientiane at the time, a hundred and fifty kilometres or so to the south, and the weather where I was did not look like a catastrophe. It looked like rain.
The water arrived anyway. Not in an hour and not in a wall. It arrived the way a schedule arrives, over days, from places I could not see, and it put parts of the capital under it.
That is the thing about river flooding that no amount of watching the sky prepares you for. The rain that floods you very often did not fall on you.
The rain that floods you fell somewhere else
Look at where Laos sits and the mechanism becomes obvious, though it is not obvious at all while you are standing in it.
The Mekong runs the length of the country from the north. Luang Prabang sits on it. Vientiane sits on it, several hundred kilometres downstream. Everything that falls on the mountains of the north has one place to go, and it goes there by gravity, at a speed set by the river rather than by the weather.
The 2023 event has the numbers attached to it. Heavy monsoon rain began on 4 August. Between 1 and 8 August the Mekong at Vientiane rose four and a third metres, an increase of roughly forty per cent in a week, and the Mekong River Commission was forecasting that it could reach twelve and a half metres. For scale, the 2018 flood at the same location peaked at 11.67 metres and left riverside shops almost completely submerged.
By the time the government finished counting on 21 August, the flooding and landslides had affected 22,024 families, some 110,120 people, across 550 villages in 50 districts of 12 provinces. Six people died and one was missing. And the provincial list is the part that tells the story: Luang Prabang in the north, Vientiane Province in the middle, Vientiane Capital at the bottom. That is not three separate disasters. That is one body of water working its way downhill through a country, arriving in each place on its own schedule.
There is a second path, and this is the one that connects Vang Vieng to Vientiane specifically. The Nam Song, the river Vang Vieng is built along, feeds the Nam Ngum system. The Nam Ngum runs three hundred and fifty kilometres out of the mountains of Xiangkhouang province, south through Vientiane province, and joins the Mekong at Vientiane itself. The two are linked by the Nam Song diversion dam.
So the capital does not receive one flood. It receives a confluence. Water coming down the Mekong mainstem from the far north, and water coming down the Nam Ngum from the northeast, converging on one city on two separate timetables. When Vang Vieng was under water and people were using boats, that water was not going to disappear. It was already on its way to me.
A storm has a track, but a river has a timetable
This is why storm forecasting, which people are now reasonably good at reading, does not help you here.
A tropical cyclone is a moving object with a position and a heading. You can put it on a map, draw a cone around it, and say roughly when and where it will arrive. Everything in the public warning system is built around that shape: watches at forty-eight hours, warnings at thirty-six, evacuation orders keyed to landfall. The whole apparatus assumes the dangerous thing is coming toward you through the air and that you can see it coming.
River flooding does not have that shape. There is no landfall. There is no cone. The dangerous thing is already on the ground hundreds of kilometres away, converting into a volume of water that will pass your town on a schedule fixed days ago by rain you never experienced. By the time you can see the problem, the problem is not preventable, only survivable.
The clearest illustration of that gap is what happened at the riverside night market in Vientiane on 9 August. Chao Anouvong Park sits on the bank of the Mekong, and below the market there is a small fairground with a Ferris wheel and bumper cars. As the river came up, workers were still down there pulling electrical cables out of the rides, trying to get them clear before the water reached them. It did reach them. The rides went under.
The Department of Hydrology and Meteorology had issued a flood warning. The vendors did not experience it as a warning. One of them, a woman selling dart balloons, was woken at three in the morning to move her stall. Others said flatly that nobody had told them, that the water simply rose faster than anything they had been given time to act on. That is the characteristic failure of this kind of flooding. The information existed. It did not travel at the speed the water did.
What that changes practically is what you watch. In storm country you watch the forecast. In river country you watch gauges. The Mekong River Commission publishes water levels at stations along the river, Luang Prabang among them, and a number rising at a station upstream of you is a more reliable statement about your next three days than any amount of looking out of the window. Almost nobody does this. I did not, the first time. I checked the weather, saw rain, and drew the wrong conclusion about what rain meant.
There is a second reason the timing catches people out, and it is about ground that has already had a month of rain on it. Early in the wet season, soil absorbs. Rain falls, the ground takes a large fraction of it, and rivers rise modestly. By August that capacity is gone. The same rainfall that produced a manageable rise in June produces a much larger one, because almost everything that falls now runs straight off into the nearest stream. This is why the destructive floods cluster in the back half of the season, and why a system that would have been unremarkable in June is the one that puts a town under water in August. The storm is not bigger. The ground is full.
Defences built by hand, out of the ground
The part I did not expect was the response.
When the water was coming, the government had people building embankments. Not concrete, not prefabricated barriers, not the machinery you see in flood footage from wealthier countries. Earth. People shifting soil into a bank between the river and the town, by hand, while the river rose.
It works in the sense that anything holding back water for a few days works, and it is also fragile in exactly the way you would expect. During the same 2023 flood, a section of the Mekong embankment at Vientiane gave way and put low-lying farmland under water. That is what a hand-built defence does when the river stays high longer than the bank was ever meant to hold.
It is an honest picture of what flood defence looks like in a country with limited money and a very long river. Laos is landlocked, low-income, and drained by one of the largest river systems in Asia. Permanent engineered defence along that entire length is not a budget question with an available answer. So the defence is temporary, improvised, labour-intensive, and rebuilt every year.
I would encourage anyone reading this from a country with functioning flood infrastructure to sit with that image for a second before judging it. That earth bank is not incompetence. It is the available option, deployed fast, by people who knew the water was coming because the towns upstream had already told them.
The water goes down, which is not the same as it being over
Here is the part that stayed with me longest, and it is the part that never makes the news cycle.
The flood ends. The river drops back inside its banks, the roads reappear, the mud gets cleared. And for the people whose houses had been under water, essentially nothing followed. No compensation scheme, no reconstruction programme, no meaningful assistance for the neighbourhoods that had been submerged. The water left, and the households were left with whatever it had done to them, and that was the end of the matter.
The scale of what goes unaddressed is in the same government report. That August, the floods damaged more than 183,000 hectares of agricultural land, 67 roads, 67 irrigation systems, three district hospitals, three schools and 110 vehicles. Agricultural land is not an abstraction in a country where a large share of households farm. A destroyed crop is a year of income, and an irrigation system that is not rebuilt is the next year as well.
That is a structural fact rather than a moral failing, and the arithmetic is not hard to follow. Recovery funding is expensive, floods in the lower Mekong are annual rather than exceptional, and a country with a small tax base cannot underwrite an annual event across its entire river frontage. But knowing why it happens does not change what it looks like from inside a house that had a metre of river in it.
It also changes what the flood actually costs. Where recovery funding exists, a flood is a bad few weeks. Where it does not, a flood is a permanent step down for the households it touched, and the next one arrives before the last one has been absorbed. The visible event is the water. The real event is what does not happen afterwards.
If you are travelling in Laos during the wet season, the practical version of all this is short. Rain where you are is not the information you need. The rivers upstream are. Luang Prabang, Vang Vieng and Vientiane are on connected water, and trouble in one is a forecast for the others with a lag of a day or two. Build slack into travel plans between June and October, because roads close and they do not reopen on your schedule.
And if a guide tells you today is not happening, that is not caution. That is somebody who has already seen what today turns into.
The storm was never over us. The water still came.
Where to check this yourself
River flooding is watched with gauges rather than forecasts, so these are the places to look. The figures in this article come from them.
- Mekong River Commission — Luang Prabang station — live water level upstream of Vientiane. This is the number that tells you about your next few days.
- Mekong River Commission — basin-wide flood forecasting and the source of the 2023 river-level figures.
- ReliefWeb: Lao PDR floods and landslides, August 2023 — the official damage and casualty totals cited above.
- Laotian Times, 9 August 2023 — the contemporaneous report from the Vientiane riverside night market.
Comments
Post a Comment