Floods · 17 to 22 August 2026 · published 30 August 2026

Rain fell across the Yungas, and every river Bolivian reporting named reached its highest in 48 years.

A flood was declared here, and we dated the peak before reading any of it.La Asunta declared a municipal disaster on 2026-08-22 under Ley Municipal 17/2026, naming the Rio Boopi and citing damage to homes, crops and roads across eleven communities, with a municipal damage report dated 2026-08-20. A separate flood was reported at Quime on the Rio Kuyuma. Both rivers rank first of 48 years in our discharge data for 17-22 August. We put the peak on 19 August from rainfall and a routing model. The municipal damage report is dated 20 August. The declaration is at La Asunta on the Boopi; our flagged cell is a model cell on the same river system 42 km away. Same catchment, not the same point. We have not verified the town's coordinates against a gazetteer, and we have not counted anyone affected.

Seven towns were named in Bolivian reporting three days after the water came. Checked against their own 47-year records, all seven rank first. The reporting chose the places; we did not. Rainfall is measured; river discharge is MODELLED, not observed.

Caranavi is the one to look at. 16 mm of rain fell there and its river ran at 621 m³/s, 4.2 times its own median, because it sits downstream and collects from the catchments above it. The water did not fall on Caranavi.

Rain, then river

Rain above, river below, one day apart. 40.3 mm fell on 18 August. The next day the river reached 1,025 m³/s, against a quiet level of 6.2. Discharge is modelled, not gauged.

Every place the reporting named

Place	Discharge	Against its own median	Rank	Rain, 17 to 22 August 2026
La Asunta	951	15.9×	1 of 48	67 mm
Caranavi	621	4.2×	1 of 48	16 mm
Chulumani	225	8.6×	1 of 48	110 mm
Coroico	147	4.7×	1 of 48	80 mm
Irupana	126	7.8×	1 of 48	110 mm
Yanacachi	72	4.4×	1 of 48	65 mm
Quime	30	38.5×	1 of 48	51 mm

Discharge is Copernicus GloFAS v4 (LISFLOOD): a hydrological model routing water downhill, forced by reanalysis precipitation. Not a gauge.

Regionally there is nothing to see, and that is the point. Across 9,376 river cells in Peru, Bolivia and northern Chile over the same days, 621 rank first against their own records and 5,363 sit below their own median. A region is the wrong unit for a flood that happened in a valley.

The international alert was 1,128 km away, on the other side of the divide. On the same days the Lima coast took 0.39 mm against a median of 0.84, which is 21 of 27 years: drier than usual rather than wetter. Nothing we measure supports a flood there in this window.

What this does not say

Nobody has been counted. We do not know how many people were affected, how many homes, or how much land. This page measures rainfall and models river discharge; neither counts anyone.

The declaration is at La Asunta on the Boopi; our flagged cell is a model cell on the same river system 42 km away. Same catchment, not the same point. We have not verified the town's coordinates against a gazetteer, and we have not counted anyone affected.

One email a week

We find climate signals in the data, and send you the ones that matter.

Confirmation email required. No spam, and every figure here stays free and public whether you subscribe or not. What you get, and how often →