On the morning of August 26, the sky above Rasuwa in northern Nepal was clear. There was no rain and no warning. High above the Nepal-China border, a mass of ice and rock broke away and fell into the Lhende Khola, blocking the river for a few minutes before the blockage gave way.
The surge that followed carried enough force to register on seismometers worldwide as a magnitude 5.2 event. The US Geological Survey later confirmed that the tremor felt in the valley was not an earthquake. It was the collapse itself.
What came next took minutes, not hours. Scientists estimate the wave covered the first 22 kilometres at an average speed of around 193 kilometres per hour. The Trishuli rose by as much as nine metres in 30 minutes. Not nine centimetres. Nine metres, in the time it takes to drink a cup of tea. At the time of writing, Nepal’s National Disaster Risk Reduction and Management Authority had confirmed 1,114 deaths, with roughly 4,500 people still missing on both sides of the border and close to 12,000 rescued. Both numbers will almost certainly have moved by the time this is read. It is already the deadliest disaster Nepal has suffered since the earthquake of 2015. Thirty-two bridges and around 40 kilometres of road have gone, along with hydropower plants and the Rasuwagadhi border crossing that links Kathmandu to Tibet.
Nepal was not without instruments. A monitoring station near the border was transmitting river data every ten minutes, and it did so normally until 8:40am. The 8:50am reading never arrived, because by then the station no longer existed. Word reached the Department of Hydrology and Meteorology roughly 16 minutes after the water had already crossed into Nepal, and the text alerts that followed went out to communities the flood had already passed through. The system worked as designed. It was simply designed for a monsoon river rising slowly, not for a mountain falling into it on a dry morning.
That is not a Nepali failure. It is a South Asian one, and Pakistan should read it as a letter addressed to us. Consider where the water went. The Trishuli joins the Narayani, which crosses into India as the Gandak. Within a day, India was evacuating people from low-lying areas of Bihar, Uttar Pradesh and Uttarakhand, and Bihar had opened all 36 gates of the Gandak barrage at Valmikinagar. Bodies were later recovered some 240 kilometres downstream inside India. A glacier failed on the Tibetan frontier and farmers in Bihar left their homes. What happened in Nepal did not stay in Nepal, and it never will.
The Hindu Kush Himalaya is the water tower for close to two billion people, and its glaciers lost ice around 65 per cent faster between 2011 and 2020 than in the decade before. As the ice retreats, slopes destabilise, the permafrost that cements the mountainsides thaws and meltwater collects in places nobody has mapped. The hazard is transboundary. The response remains stubbornly national.
Pakistan holds more glaciers than any country outside the polar regions, and more than 3,000 glacial lakes sit across Gilgit-Baltistan and Khyber Pakhtunkhwa. The UNDP classifies 33 of them as hazardous, with around seven million people living below them. In 2025, glacial lake outburst floods killed dozens in Gilgit-Baltistan and cut off valleys such as Diamer and Ghizer. Look at the geography of Rasuwa and it is our own: a narrow valley, a road laid along the riverbed because there was nowhere else to put it and villages sitting under a lake that did not exist when their grandparents settled there.
At the other end of the country, the water is arriving permanently. I have stood at Keti Bundar and looked out across open sea towards where the village used to be. This is not a projection in a report. Vast tracts of Thatta, Sujawal and Badin have already gone to saltwater. Drinking water arrives by drum. Fishing grounds have collapsed, and the mangroves that once absorbed the storm surge have thinned. I met families who had moved twice already, and fishermen who now work other men’s boats because they lost their own.
Ask where those families go and the answer is Karachi. They end up in the katchi abadis on the edges of the city, working as labourers and domestic staff, in settlements with no security of tenure and no services. Nobody counts them. No register records people displaced by seawater intrusion, no category exists for them in official statistics and no line item appears for them in any provincial budget. They are quietly absorbed into the urban poor, and the reason for their poverty disappears from the record. If we cannot count them, we cannot plan for them, and we will keep treating what happened to them as bad luck rather than policy failure.
In my research on climate displacement in India and Pakistan, I have called this ‘compound displacement’. Not one dramatic exodus, but repeated movement that hardens into permanent precarity. A household moves after the flood, returns, rebuilds on borrowed money, loses the next crop to salinity, and by the third move it has no assets, no documents and no way back. In 2022, around eight million Pakistanis were driven from their homes. In 2025, the monsoon floods killed more than a thousand people, affected close to seven million and forced some three million to flee. What those figures hide is that these are frequently the same families.
Displacement never travels alone. It reaches into health, because displaced people get waterborne disease and untreated illness. It reaches into education, because girls pulled out of school rarely return. It reaches into our cities, because people arriving on the fringes of Karachi and Lahore meet no housing policy, only the next floodplain. The atmosphere is not the crisis. The crisis is what a changed atmosphere does to societies that were already fragile.
There is also no legal home for these people. Pakistan has no law on internal displacement, no protected status for those displaced by disaster, no resettlement framework and no enforceable claim against anyone. Our climate policy documents are full of the word resilience and empty of rights.
We all know the pattern by now: something breaks, the pictures are unbearable, money is pledged and the talk shows fill for a fortnight. Then the water recedes, the cameras leave and the file closes until the next one. Nepal knows it too. The same river flooded in July 2025 and destroyed the Nepal-China Friendship Bridge at the same crossing, and the warning gap identified then was still open on August 26. We are no better. We commission a report after every flood and implement the cheapest recommendations in it.
So, what would seriousness actually look like? Hazard mapping and real monitoring for every glacial lake sitting above an inhabited valley, not the handful of pilot projects we currently fund. Early warning built for sudden events, with sensors placed to survive the very thing they exist to detect, and alerts that reach a shepherd’s phone in Ghizer rather than only a control room in Islamabad. Siting rules that treat a riverbed as a riverbed, and the will to enforce them against those who build in it anyway. A national register of the displaced, so that the people leaving the delta stop being invisible. And a legal framework that gives a family washed out of Keti Bundar or Ghizer a status and an entitlement rather than a tent.
None of this is expensive compared with what one bad morning costs. Nepal has handed us a warning that has cost more than a thousand lives, and it has cost us nothing at all to receive. The only question is whether we act before our own August 26 or after it.
The writer is an environmentalist.
Disclaimer: The viewpoints expressed in this piece are the writer’s own and don’t necessarily reflect Geo.tv’s editorial policy.
Originally published in The News
2026-09-05 09:02:00
