Expedition · Cities
Dhaka floods when it barely rains
The city has drainage. It has pumps, canals and a masterplan with a number on it. What it does not have is anywhere for the water to go once it arrives — and that is a choice someone made, slowly, over forty years.
There is a particular kind of afternoon here where the sky opens for twenty minutes, stops, and the road outside your building is still a river at nine that evening. Twenty minutes. Not a cyclone, not a monsoon record — a rain that a city of twenty million should absorb without anyone reaching for their phone camera.
I went looking for the reason, expecting to find a broken thing. A pump that failed, a budget that vanished, a drain nobody cleaned. What I found instead was a system doing exactly what it was built to do, in a city that stopped resembling the one it was built for.
The water used to have somewhere to go
Every city is a bargain with water. You cover the ground in something hard, the rain stops soaking in, and you owe that water a route. Dhaka’s original route was not concrete — it was the low ground: a lattice of khals and the wetlands they drained into. Rain fell, ran downhill, sat somewhere flat for a while, and left.
That lattice is mostly gone. The water did not get a memo. It still runs downhill, still arrives at the same low ground — and now finds a four-storey building where the wetland was.
You can pump water out of a bowl. You cannot pump it out of a bowl that is filling faster than the pump, into a river that is higher than the bowl.
Three things have to be true at once
Urban flooding gets reported as one failure. It is almost always three, stacked. Pull any one of them out and the street stays dry.
- The rain arrives faster than the ground can take it. Not more rain overall — more rain per hour, on more concrete.
- The route out is narrower than it was. A canal squeezed into a pipe carries a fraction of what it used to, and blocks on things a canal would have carried away.
- The destination is full. During monsoon the rivers around the city sit high. Water does not flow uphill because the drainage plan would prefer it to.
The third is the one that gets left out of the conversation, and it is the one that turns a drainage problem into a hydrology problem. It is also the one no amount of cleaning the drains will fix.
What a retention pond actually does
The unglamorous fix is a hole. A retention pond does not remove water — it borrows time. It holds the peak of the storm for the two hours when everything downstream is full, and lets it out afterwards, when the river has dropped and the pipe is free. The engineering is trivial. The hard part is that a hole in the ground is worth more with a building on it, and the benefit of the hole is invisible on every day it does not rain.
This is the shape of nearly every climate-adaptation argument, and it is worth naming plainly: the cost is now, visible, and paid by someone specific; the benefit is later, invisible, and spread across everyone. That asymmetry is not a Dhaka problem. It is why Houston floods, why Jakarta sinks, why the same conversation is happening in Chennai this week.
What you can see from a rooftop
You do not need a hydrological model to read this city. Stand on any rooftop in Mirpur after rain and you can see the old water map in the shape of what is wet — the low ground remembers, even where it has been built on. Someone paved a river and put a bus route on it.
- Water pools where the map used to be blue, not where the drains are worst.
- The oldest neighbourhoods on high ground rarely flood, and nobody planned that either.
- New development sits on the flattest, cheapest land — which is flat and cheap because it was wetland.
The part I got wrong
I started this thinking the villain was corruption, because that is the easy story and there is enough of it to make the story stick. It is a real factor and I am not waving it away. But a city can be perfectly honest and still end up here, because every individual decision — fill this plot, culvert this canal, build on this field — was locally reasonable, and the damage only shows up in aggregate, twenty years later, on an afternoon when it barely rained.
That is a harder problem than a villain. It is also a more useful one, because it means the fix is institutional rather than moral: someone has to be responsible for the aggregate, and be able to say no to a locally reasonable thing.
Next issue
I want to follow the water out — to the pumping stations at the edge of the city, and the question of what happens when the river they pump into is the thing that is rising. If you know this system from the inside, I would genuinely like to hear where I have this wrong.
— Tuna
Notes
- Specimen text. The figures and the quotation are invented for the purpose of setting type and should not be read as reporting.
- Inline reference style: a link inside body copy.