Borrowing a term from dam engineering, this can be thought of as a “blue-sky flood”: a dangerous flood generated event when the weather itself gives little indication of what is about to happen. The mountain produced a detectable signal before the destructive flood reached communities downstream, but the challenge was that this signal was not connected to the conventional flood-warning system.
Hossein Bonakdari, Associate Professor, at uOttawa’s Department of Civil Engineering, explains what happened and what it means for disaster preparedness.
Q: What actually happened?
Bonakdari: The event appears to have started at approximately 5,200 metres elevation, where a large mass of ice and rock detached and fell roughly 1,200 vertical metres into the Lhende Khola system. But the initial collapse was only the beginning. As the material accelerated downhill, it collected additional rock, sediment, ice and water. The moving mass transformed into a fast, debris-rich flow and appears to have temporarily blocked the river. The sequence can be understood as:
Ice and rock collapse → rapid movement downhill → debris entrainment → river blockage → temporary natural dam → dam failure → destructive flood wave.
This distinction is important. What travelled downstream was not simply “a collapsing glacier.” It was a cascade of connected processes involving ice, rock, debris and water. Each process amplified the next.
Q: Was the disaster caused by an earthquake?
Bonakdari: The available evidence indicates that it was apparently not a tectonic earthquake. This is one of the most important aspects of the event. A seismic signal initially interpreted as an approximately magnitude-4.4 earthquake was subsequently reassessed as having been generated by the mass movement itself, with the collapse producing a signal approximately equivalent to magnitude 5.2. Put simply:
- The mountain moved strongly enough to look like an earthquake.
- That distinction has major implications for disaster warning.
- The seismic signal appears to have been a signature of the collapse, rather than the cause of it.
- And that means the event was detectable.
The critical problem is that detecting an event and issuing a flood warning are not currently the same system.
Q: Why does the seismic signal matter for early warning?
Bonakdari: Conventional flood-warning systems are designed primarily to monitor rainfall and rivers.
They ask questions such as:
- How much rain is falling?
- Is the river rising?
- How saturated is the watershed?
- How much water is moving downstream?
Those measurements are extremely useful for conventional rainfall-driven floods. But this event began high in the mountains and outside the variables normally monitored by a flood-warning system.
- The weather could look normal.
- The river could still look normal.
- Yet a major collapse could already have occurred upstream.
This suggests an important change in how we think about warning systems in high-mountain regions: we need to monitor the mountain as well as the weather and the river.
Q: Did climate change cause the collapse?
Bonakdari: This question requires scientific caution. We should not say that climate change has been demonstrated to have caused this specific collapse. Establishing that connection requires detailed scientific analysis. However, climate change is changing the environmental conditions in which these events occur. Across the Hindu Kush Himalaya, warming is affecting glaciers, snow, frozen ground, meltwater and high-mountain lakes.
For example:
- Retreating glaciers can remove support from steep valley walls.
- Warming can reduce the stabilizing effect of ice within fractured rock.
- Meltwater can increase water pressure within slopes and contribute to unstable lakes.
- Glacier retreat can expose large amounts of loose sediment that can be carried downstream during a collapse.
So, the scientifically appropriate message is not: “Climate change caused this collapse.” It is: “Climate change is changing the conditions in which these failures can occur.” That distinction is important.
It allows us to recognize the growing risks associated with a warming high-mountain environment without attributing a specific disaster to climate change before the evidence supports that conclusion.
Q: Could something similar happen in other mountain regions?
Bonakdari: The broader lesson extends beyond Nepal. High-mountain environments in the Himalayas, Alps, Andes and other mountain regions contain glaciers, steep slopes, frozen ground and lakes that can change as temperatures rise. The precise mechanism will differ from one location to another, but the fundamental concern is similar: ice and rock can become unstable, large collapses can occur, rivers can be temporarily blocked, and sudden releases of water and debris can threaten communities downstream. This is why the Nepal event should be viewed not only as a local disaster, but also as an important warning for other high-mountain regions.
Q: What should a modern early-warning system monitor?
Bonakdari: There is no single sensor that can solve this problem. What is needed is an integrated, multi-sensor warning system that combines different types of information.
1. Seismic monitoring: Seismic instruments can detect the ground vibrations produced by very large ice and rock collapses. Automated systems could help distinguish these signals from ordinary earthquakes.
2. River monitoring at high altitude: River gauges farther upstream could detect sudden changes in water level or flow and automatically transmit alerts downstream.
3. Satellite monitoring: Satellite images can help identify changes in glaciers, growing lakes, new natural dams and other changes in high-mountain environments.
4. Slope-movement monitoring: Satellite radar, including InSAR, can help identify gradual movement of some unstable slopes before failure.
5. Ground-based sensors: GNSS and other instruments can be installed at selected high-risk locations to measure movement directly.
6. Automated data integration: This is perhaps the most important part. Seismic signals, slope movement, glacier changes, lake changes and river levels should not remain isolated in separate databases. The sensors need to communicate with one another through a common warning system capable of recognizing danger and triggering an alert quickly.
7. Cross-border information sharing: The watershed does not recognize the Nepal–China border. Neither should the warning system. Hazard information from upstream areas must move across borders faster than the flood itself.
Q: What is the main lesson from this disaster?
Bonakdari: The central lesson is that the absence of rain does not mean the absence of flood risk, and the absence of an earthquake does not mean the mountain is stable. Our flood-warning systems have become very good at monitoring the sky and the river.
In high-mountain environments, we increasingly also need to listen to the mountain. Seismic waves can travel through the Earth in seconds. A destructive flood wave travels downstream much more slowly. That difference creates a potentially valuable warning window. The engineering challenge is therefore straightforward: Detect the failure quickly. Understand what the signal means. Estimate where the hazard is going. And get the warning downstream before the flood arrives. In disaster warning, information has to travel faster than the hazard.
---
For interviews, media may contact Professor Hossein Bonakdari at [email protected] .