What a 132-yr-old Himalayan disaster can teach us about Nepal flash floods

The devastating flash flood that struck Nepal’s Rasuwa district on the morning of Wednesday, 26 August was a reminder of how quickly the Himalayas can turn from a source of life into an agent of destruction. Water, sediment and huge boulders raced through the Bhote Koshi and Trishuli river systems, devastating settlements and infrastructure.

Preliminary assessments by scientists suggest that an ice-rock avalanche may have fallen into the Lhende Khola, a tributary of the Bhote Koshi, triggering the sudden surge.

The speed was extraordinary. At Galchhi, downstream on the Trishuli, the river level reportedly rose by as much as nine metres in just 30 minutes. Scientists are also examining whether avalanche debris temporarily blocked the river, creating the possibility of another sudden release downstream.

There is little humans can do to prevent such geological events. But they can understand the hazard, monitor it, communicate warnings and move people out of harm’s way.

That lesson was demonstrated spectacularly — not by satellites or artificial intelligence, but by engineers, geologists and administrators more than 130 years ago in the Garhwal Himalayas.

On 21 September 1893, a massive landslide blocked the Birehi Ganga, a tributary of the Alaknanda, in present-day Uttarakhand. The landslide came down from a high mountain slope, creating a gigantic natural dam. Behind it, water accumulated to form what became known as Gohna Lake.

The scale of the event was enormous. Geological accounts estimate that millions of tonnes of rock collapsed into the valley, creating a lake several kilometres long and hundreds of metres deep.

The authorities understood almost immediately that the lake itself was a threat. If the natural dam failed, an enormous volume of water would rush downstream through the Birehi and Alaknanda valleys, potentially reaching Srinagar and beyond.

What followed was one of the earliest remarkable examples of organised disaster-risk management in the Himalayas. Army engineer Lieutenant Crookshank and engineer Thomas Holland were sent to monitor the lake. A temporary telegraph line was established between Gohna and Chamoli so information about the lake could be transmitted rapidly. Water levels were recorded regularly and engineers studied the behaviour of the unstable natural dam.

The authorities did not stop at observation. Posts were established downstream and danger levels marked along the Alaknanda and Ganga valleys. People living in vulnerable areas were instructed to evacuate when the river crossed specified levels. Suspension bridges were dismantled, pilgrim routes diverted and villages in the anticipated path of the flood evacuated.

Most remarkably, engineers tried to predict when the natural dam would fail. On 22 August 1894, Crookshank warned that the flood was likely to begin within two days. The prediction proved remarkably accurate. Around midnight on 25-26 August, the lake overtopped the landslide dam and the barrier partially collapsed, releasing a huge surge downstream.

The destruction was immense. Srinagar, then an important town in Garhwal, was devastated. Fields, houses, bridges and other infrastructure were swept away. Contemporary estimates put the volume of water released at more than 10,000 million cubic feet.

Yet the most important statistic was not the volume of water released or the amount of property destroyed.

It was the number of people killed. Almost nobody died.

Contemporary accounts noted that warnings had been transmitted down the valley and people moved to safety. One later account records five deaths in a family that ignored the evacuation warning; other historical accounts describe the event as causing no reported loss of life.

Gohna is not a story about defeating nature. The landslide could not be prevented. The lake could not simply be wished away. When the dam finally failed, the flood could not be stopped. What could be controlled was exposure.

That is the crucial distinction for the Himalayas today. Gohna Lake survived in a reduced form for another 76-77 years. In July 1970, another massive flood in the Alaknanda basin filled the lake with sediment and contributed to its final destruction. The 1970 disaster killed hundreds of people and caused widespread destruction across the valley.

The contrast between 1894 and 1970 is instructive. The Himalayas remained just as unpredictable. What changed were the nature and scale of the event, conditions on the ground and the vulnerability of people and infrastructure.

Today, we have technologies the administrators of 1894 could scarcely have imagined. Satellites can repeatedly scan inaccessible mountain slopes. Seismic instruments can detect sudden movements. River gauges can transmit water-level data in real time. High-resolution imagery can identify landslides, unstable slopes and changing glacial conditions. Scientists can model how a blockage might translate into a flood wave hundreds of kilometres downstream.

But technology is useful only when it is converted into action. The question is whether we have built systems capable of turning that technology into precious minutes of warning — and those minutes into lives saved.