Nepal, the mud of Rasuwa a week later: the collapsed glacier and the buried villages

The starting point is a date and a watercourse: on August 26, 2026, a sudden flood swept along the Bhote Koshi river, from the Tibetan slope down into Nepal’s Rasuwa district. At its origin, according to the reconstruction reported by CNBC, was the collapse of a glacier, which triggered flash floods on both sides of the border and caused hundreds of deaths. A confirmed death toll does not currently emerge from the available material: the figure that can be published remains an order of magnitude, and should be read as provisional.

The second verifiable element is temporal, and it says more than many descriptions could. Aerial images from September 3 — eight days after the wave — show the homes of the village of Betravati, in Rasuwa, still surrounded by mud, rocks and debris. This is not water receding: it is solid material that stays where it has landed, and that must be removed before reconstruction can even be discussed. It is the feature that distinguishes this type of flood from an ordinary lowland flood, and that extends the emergency well beyond the hours of peak impact.

The context in which the event is set is documented: the same CNBC review places Rasuwa within a series of natural disasters recorded over the past month, from earthquakes to droughts to typhoons. It is the overlap that matters, more than the single episode: multiple emergencies in close succession compete for the same relief resources and the same international attention, and an event confined to mountain villages along a single river is the one most likely to fall out of the news cycle while the debris is still on the ground.

It is worth considering what a disaster of this scale means, in terms of accounting, once estimates arrive. A possible comparison comes from Colombia, and it is material to be handled with an explicit caveat: on that episode, the available data trace back to a single source, the Turkish state agency Anadolu, picked up by CNBC; there is currently no independent confirmation. For this reason, the figures that follow are not published as established facts, but as an attributed statement. According to Anadolu, on August 10, 2026, an earthquake of magnitude 7.4 struck western Colombia with its epicenter in the Pereira area; the reported toll is 289 dead, 4,187 injured and 143 missing; on August 18, President Abelardo de la Espriella reportedly released a preliminary estimate of economic losses amounting to $9.57 billion. Still according to the same source, fourteen days after the quake aerial footage over Pereira still showed the rubble of collapsed buildings.

The parallel is not meant to add up tragedies, but to isolate a measurable constant, with the difference in evidentiary status already noted: the Nepalese case rests on a journalistic review, the Colombian one on a single state agency. In both cases the visible phase of the event lasts minutes — the tremor, the wave — while the documentable phase lasts weeks, and can be read in the persistence of debris in images taken days apart. In both cases the tolls released in the immediate aftermath are estimates: those on casualties may move in either direction as searches continue, and the Colombian economic figure is presented as preliminary by the very statement that contains it.

What remains to be noted is what cannot yet be verified about Rasuwa. There is, from the available material, no official estimate of economic damage in Nepal, no definitive count of the villages affected beyond Betravati, nor any breakdown of casualties between the Tibetan and Nepalese sides. The solid data point, as of today, is the images from September 3: in Rasuwa, mud and rocks were still pressed against the houses.

Sources: CNBC; CARE (humanitarian organization). Data on Colombia are attributed to Anadolu (Turkish state agency), picked up by CNBC.