Nepal’s Disaster Was Not Simply a Flood
How and Why It Happens: The Science Behind This Hazardous Event
What happens when a large mass of rock and ice collapses from a mountain more than 4,800 metres above sea level?
It can become much more than a landslide.
A recent disaster in Nepal showed how a collapse high in the Himalayas can turn into a powerful mixture of rock, ice, water and debris, travelling through a valley and causing devastating destruction downstream.
At first, calling it a flood seems reasonable. The reason There was a huge amount of water and debris moving downstream. But when scientists look at how the event may have developed, the picture becomes much more complicated and different from earlier assumptions.
The exact sequence is still being investigated.
One possibility is that a large bedrock landslide carried the glacier with it. Another is that the bedrock landslide first destabilised the glacier, causing it to collapse shortly afterward. It is also possible that the glacier failed first and the bedrock above it followed.
So, what actually happened?
One current interpretation is that a rock-ice avalanche began at around 4,800 metres and travelled more than 1,200 metres downslope.
The involvement of ice may have changed everything.
As the rock and ice broke apart during the collapse, fragmentation and friction generated heat. Some or much of the ice could melt during this process, adding water to the moving material.
Now imagine a mixture of broken rock, ice and water moving rapidly downhill.
Instead of behaving like a dry rock avalanche, it can become a much more fluid and mobile debris flow. This can allow it to travel much farther downstream, making it a long-runout debris flow.
But the process doesn't stop there.
As the flow moved along the river, it could erode the valley floor and sides and pick up more material. This process is known as entrainment. More water, sediment and rock can therefore become part of the moving flow as it travels downstream.
Farther downstream, when the flow loses energy, the material is deposited across the landscape.
So the disaster may be better understood as a chain of processes:
rock and ice collapse → rock-ice avalanche → fragmentation and friction → ice melting → water-rich debris flow → erosion and entrainment → downstream deposition.
This is why some experts describe the event as a hazard cascade or process chain.
One hazard develops into another.
But was it a Glacial Lake Outburst Flood?
Early reports described the disaster as a possible Glacial Lake Outburst Flood.
It's easy to understand why.
GLOFs are a recognised hazards in the Himalayas, where glacial lakes can store large amounts of water and suddenly release it downstream.
But a GLOF specifically involves the sudden release of water from a glacial lake.
According to the scientific material available for this event, there was no lake involved.
That difference may sound small, but scientifically it matters.
In a GLOF, scientists can identify the lake, estimate how much water it contains and broadly model where the flood could travel.
A rock-ice avalanche that transforms into a debris flow is different. The hazard can change as it moves.
The source is a collapsing slope rather than a known body of stored lake water. Ice can melt. Water can be generated. The flow can erode its surroundings and entrain more material.
By the time it reaches downstream areas, it may look very different from what started the process high in the mountains.
That's why hazard cascade or process chain may describe this event better than simply calling it a GLOF.
Where does climate change come into this?
This is probably one of the most difficult questions to answer without oversimplifying the science.
Across the Himalayas, glaciers are generally thinning and retreating. This can leave behind glacial lakes and slopes that may be unstable.
Another factor is permafrost.
Permafrost is permanently frozen ground, and in high mountain environments it can help hold rock and slopes together. When it thaws, it can affect slope stability.
So, climate change can influence the conditions in which mountain hazards develop.
But that doesn’t mean we start saying climate change directly caused this particular disaster.
There are other factors that can precondition a mountain slope for failure.
Earthquake damage can weaken rock. The Himalayas are extremely steep. The mountains are also rapidly uplifting, while gravity is constantly working in the opposite direction.
These factors may interact.
Climate change may play a role, potentially an increasing one, but the exact contribution to this particular event is still something researchers are investigating.
And there is another part of the story that is sometimes missed.
The Himalayas aren't empty mountains.
People live in these valleys. They build homes, grow crops and develop infrastructure. Hydropower is also important in these landscapes.
This means that when a mountain hazard occurs, the physical event is only one part of the disaster.
The other part is exposure.
Why did the event become so destructive?
A large amount of rock, mud and water moving rapidly through a valley leaves less time to react.
Anything on the valley floor can be exposed. The flow can also move around bends and wash against lower parts of the valley sides.
This is one reason the same type of mountain failure can have very different consequences in different parts of the world.
Rock avalanches also occur above glaciers in places such as Alaska. Some can travel several kilometres.
Get Sittara Shaikh’s stories in your inbox
Join Medium for free to get updates from this writer.
But if the material doesn't become fully fluid, its behaviour is different. And if people aren't living immediately downstream, the consequences can also be very different.
In the Himalayas, steep and retreating glaciers can be relatively close to communities.
That changes the risk.
A hazard becomes a disaster when it exceeds our ability to cope with it.
The mountain doesn't have to collapse directly onto a settlement. The material only has to reach a place where people, buildings, roads, farms or other infrastructure are exposed.
Why is it so difficult to predict?
This is where the problem becomes even harder.
There are thousands of potentially unstable slopes across the region.
But how can scientists know which slopes are likely to fail?
Simply placing instruments on every mountain slope and waiting for something to happen is not enough.
And so far, researchers have not identified obvious precursors that would have clearly warned that this particular failure was about to occur.
This makes hazard cascades from landslides much harder to manage than some other mountain hazards.
In a GLOF, scientists know where the water is coming from because it is stored in a glacial lake. They can estimate the amount of water and model where the flood could travel.
With a landslide, there may be uncertainty about when the slope will fail, how much rock and ice will collapse, how much ice will melt and how much material the flow will pick up along the way.
The hazard itself can evolve.
That makes prediction extremely challenging.
So, how do scientists understand what happened?
This is where different types of data become useful.
Satellite imagery can give researchers a wide view of a remote mountain landscape. It can help identify where the event began, where the material travelled and what has changed afterward.
Satellites can also help identify secondary hazards.
For example, debris may block water and create a new temporary lake or water buildup. Unstable material may remain on the slope and create another hazard later.
But satellite images don't tell the whole story.
Ground photographs and videos can show what the flow looked like and how the landscape changed.
Then there is seismic data.
Scientists also use seismic signals, although both landslides and floods can generate them. Researchers can study these signals to understand how quickly an event moved and distinguish landslides from earthquakes and other sources of seismic noise.
When these different sources of information are combined, they can provide a much better picture of what happened.
Remote sensing, ground observations and seismic signals each provide a different part of the story.
And perhaps this combination can do more than help us understand disasters after they happen.
It may also help us detect some events more quickly.
If seismic signals can be interpreted rapidly enough to distinguish a landslide from an earthquake, precious time may be gained.
But how much time?
For some people, perhaps only minutes.
For others, potentially longer.
That creates another difficult question: what should people actually do when the warning comes?
There isn't one simple answer.
For some locations, reaching a high concrete building may be safer. In another location, that same type of building might not provide enough protection, and moving uphill could be the better option.
The safest action depends on the location and the hazard.
This is why community preparedness has to be part of disaster risk reduction, not something considered after the monitoring system is installed.
What can we do about these hazards?
Better monitoring can help.
Targeted in-situ monitoring can provide detailed information about specific high-risk slopes. Satellite observations can give a much wider picture. Seismic monitoring may help detect rapid failures and provide additional warning.
But technology alone won't solve the problem.
People living in exposed areas need to understand the hazards around them and know what action to take when a warning arrives.
Authorities also have to consider where housing, agriculture and hydropower infrastructure are being developed.
And sometimes the difficult answer is that a particular place may simply be too hazardous to live in.
That isn't an easy decision, especially in landscapes where communities have lived and worked for generations.
But disaster risk reduction isn't only about predicting when a mountain will collapse.
It is also about understanding what happens after the collapse.
A mass of rock and ice can become an avalanche. The ice can melt.
Water can turn the moving material into a debris flow.
The flow can erode the valley and entrain more material.
And what began high on a mountain can eventually become a disaster many kilometres downstream.
We may not be able to predict every failure.
But understanding these hazard cascades can help us improve monitoring, early warning systems and community preparedness.
And perhaps most importantly, it can help us ask a difficult question before the next disaster happens:
Are we building and living in places where the landscape itself can become the hazard?
