One Flood, Three River Names: How Bhote Koshi Becomes Trishuli, Narayani and Finally Gandak

From the high Himalayan borderlands of Tibet and Nepal to Bihar’s Gangetic plains, the current flash-flood disaster has exposed an extraordinary river corridor — one in which geography, local naming, tributaries and national borders repeatedly change the identity of the water flowing downstream.

A river high in the Himalaya suddenly turns violent.

Within minutes, water mixed with ice, rock, mud and boulders races through a narrow mountain valley. A settlement hundreds of kilometres away may not share the name of the river where the disaster started. It may not even be in the same type of landscape.

But it can still be downstream.

That is one of the central lessons of the catastrophic 26 August 2026 Rasuwa flash flood.

The flood travelled through what current scientific and official reporting describes as the Bhote Koshi–Trishuli river system, with concern extending farther downstream towards the Narayani-Gandak basin. ICIMOD reported that water levels in the Bhote Koshi basin began rising rapidly at around 9 am local time before the flood moved into the Trishuli. At Galchhi, the Trishuli reportedly rose by as much as nine metres in about 30 minutes; at Malekhu the rise was around seven metres.

USGS subsequently assessed that a glacial collapse near the Nepal-China border likely triggered a catastrophic debris flow and flood, with the destructive flow travelling nearly 100 kilometres through populated areas.

The disaster therefore raises a deceptively simple question:

Are Bhote Koshi, Trishuli and Narayani actually three different rivers?

The answer is:

They are different named reaches and components of one connected downstream river system in this corridor — but they should not be treated as three interchangeable names for an unchanged river.

The river acquires tributaries, changes dimensions and assumes different geographical identities as it descends.

THE RIVER IN ONE LINE

Tibet/High Himalaya

↓

Lhende / upper Bhote Koshi reach

↓

Trishuli

↓

Devghat

↓ + Kali Gandaki and the wider Gandaki system

Narayani

↓ Nepal–India frontier

Gandak

↓

Ganga

This is the simplest way of understanding why a disaster that begins in a Himalayan gorge can eventually become a concern for communities much farther south.

1. BHOTE KOSHI: FIRST, UNDERSTAND WHICH BHOTE KOSHI

This is where much of the confusion begins.

Bhote Koshi is not an unambiguous river name in Nepal.

The term essentially carries the geographical sense of a river coming from the direction of Tibet, and more than one Himalayan river has been known by that name.

In the context of the present Rasuwa disaster, Bhote Koshi refers to the upper river corridor connected with the Lhende system near the Nepal-China border and the upper Trishuli system.

ICIMOD’s 26 August assessment explicitly described the flood as moving through the Bhote Koshi into the Trishuli and identified the Lende Khola as a stream where a large volume of ice and rock may have entered the river system.

But there is another Bhote Koshi

The famous Bhote Koshi associated with Sindhupalchok, Kodari and the Nepal-Tibet highway east of Kathmandu is another river.

That Bhote Koshi feeds the:

Bhote Koshi → Sun Koshi → Koshi

system.

It should not be confused with the Rasuwa Bhote Koshi terminology being used in connection with the present disaster.

The distinction

That single distinction resolves much of the apparent contradiction in maps and news reports.

2. WHERE DOES THE RASUWA BHOTE KOSHI COME FROM?

The upper system drains very high Himalayan terrain around the Gyirong/Kyirong region of Tibet and the Nepal-China frontier.

Streams from the Tibetan and high-Himalayan catchments descend towards Rasuwagadhi, where the river system enters the extraordinarily steep terrain of northern Nepal.

The terminology changes between local usage, historical maps and hydrological descriptions. The upper watercourses are variously associated with names including Kyirong Tsangpo, Lhende/Lende Khola and Bhote Koshi.

This is one reason assigning a neat independent length to the “Rasuwa Bhote Koshi” is misleading.

It is better understood as an upper named reach of the larger river corridor than as a completely separate river with an universally agreed beginning and ending.

Bhote Koshi/Lhende at a glance

That last point matters.

A number that looks precise can actually make the geography less accurate if one source measures Lhende, another measures the Tibetan headwater and another begins the Trishuli at a different confluence.

3. TRISHULI: THE HIMALAYAN RIVER THAT CARRIES THE SURGE SOUTH

As the upper waters descend through Rasuwa, the principal river becomes the Trishuli.

The river then runs south through or along the central Nepal corridor encompassing Rasuwa, Nuwakot, Dhading and Chitwan, while its wider basin extends across substantially more territory.

The International Finance Corporation’s basin-wide assessment describes the Trishuli River Basin as a major tributary system of the Gandaki, covering roughly 32,000 square kilometres, illustrating just how much larger the hydrological system is than the main river channel visible on a map.

Mapping datasets put the main Trishuli reach at roughly 160 kilometres, with around 163 km commonly mapped. Because river length changes depending upon which headwater is adopted as the starting point, it is safer to describe this as an approximate mapped main-stem length, not an immutable figure.

Trishuli — geography 

And it was this stretch that demonstrated the extraordinary speed of the present flood.

CURRENT FLOOD: HOW FAST DID THE WATER RISE?

Reported Trishuli gauge movement

Source: ICIMOD preliminary hydrological reporting, 26 August 2026.

Imagine a river surface moving upward by roughly the height of a three-storey building within half an hour.

That is why Himalayan flash floods are fundamentally different from the slower river flooding familiar across many plains.

The danger is not merely how much water arrives.

It is how quickly it arrives, what it carries, and how little reaction time the valley allows.

4. DEVGHAT: WHERE TRISHULI BECOMES PART OF SOMETHING MUCH BIGGER

Follow the Trishuli farther south and one reaches Devghat, one of Nepal’s important river confluence zones.

Here the Trishuli meets the larger Gandaki system.

From this point downstream, the great trunk river is known as the Narayani.

This distinction is hydrologically important.

The Narayani is not simply the Trishuli with a new road sign.

By Devghat, water from several enormous Himalayan drainage systems has become integrated into a substantially larger river.

Think of it this way

Bhote Koshi/Lhende

small, steep upper Himalayan corridor

↓

Trishuli

larger mountain river receiving tributaries

↓

Devghat + wider Gandaki waters

major confluence

↓

Narayani

large Himalayan trunk river

The farther downstream one travels, therefore, the less correct it becomes to describe the water as merely “the Bhote Koshi.”

5. NARAYANI: THE MOUNTAIN RIVER ENTERS THE TERAI

Below Devghat, the river passes through the Chitwan–Nawalpur/Nawalparasi landscape and emerges from the Himalayan foothills towards Nepal’s Terai.

The geography changes dramatically.

The steep, confined Himalayan river begins becoming a broad alluvial river.

The gradient decreases.

The channel widens.

Floodwater begins behaving differently.

Instead of being squeezed through a mountain gorge, enormous volumes of water can spread across low-lying land.

And then the river approaches India.

6. NARAYANI BECOMES GANDAK IN INDIA

Once the system crosses towards the Indian plains, the river is widely known as the Gandak.

The Government of India’s Central Water Commission has historically given the entire Gandak system a length of about 630 kilometres, of which about 380 kilometres lie in Nepal and Tibet, before the river continues across the Gangetic plains towards the Ganga.

In India the river is associated principally with the Valmikinagar/West Champaran corridor before travelling across north Bihar through districts including areas of:

West Champaran → East Champaran → Gopalganj → Muzaffarpur → Saran/Vaishali belt

before ultimately meeting the Ganga.

FROM TIBET TO THE GANGA: THE COMPLETE CORRIDOR

The key idea

The water is connected.
The river name is not constant.
And the river itself is not constant either.

It becomes progressively bigger as tributaries join it.

WHY CAN FLOODS HERE BECOME SO VIOLENT?

The Himalaya combines several ingredients that can turn an ordinary mountain river into a high-energy disaster corridor.

1. Extraordinary vertical relief

Water begins thousands of metres above the plains.

Over a relatively short horizontal distance, rivers plunge through enormous elevation differences.

That gives Himalayan rivers enormous gravitational energy.

HIGH MOUNTAINS

↓

STEEP CHANNEL

↓

NARROW GORGE

↓

HIGH-VELOCITY WATER

↓

BOULDERS + SEDIMENT + DEBRIS

↓

FLASH-FLOOD WAVE

A flood here is therefore rarely only water.

It can become a moving mixture of water, sediment, trees, rock, ice and infrastructure debris.

2. GLACIERS, SNOW AND HIGH-MOUNTAIN ICE

The upper catchments contain glaciers, permanent or seasonal snow and steep rock faces exposed to repeated freezing, thawing and gravitational failure.

USGS’s current assessment says the 26 August event was likely triggered by a glacial collapse, producing a catastrophic debris flow and flood.

ICIMOD has been more cautious about the exact mechanics: its scientists reported evidence consistent with a large volume of ice and rock entering the Lende Khola, while continuing to investigate whether temporary blockage of the river and subsequent release contributed to the surge.

So the safest scientific description at this stage is:

A major high-mountain ice-and-rock failure appears to have initiated the disaster; the exact sequence by which it generated and amplified the flood remains under scientific investigation.

3. LANDSLIDES CAN CREATE NATURAL DAMS

A landslide does not always produce its worst damage at the instant it falls.

Rock and debris can block a narrow river.

Water accumulates behind the blockage.

Then the dam fails.

SLOPE FAILURE

↓

RIVER BLOCKED

↓

WATER ACCUMULATES

↓

NATURAL DAM FAILS

↓

SUDDEN FLOOD WAVE

ICIMOD said scientists were examining precisely this possibility following the Rasuwa disaster and warned of continuing danger from upstream blockage.

This is why the danger may persist even after the first flood wave has passed.

4. THE MONSOON ADDS ANOTHER LAYER OF RISK

The Himalaya receives intense seasonal rainfall.

Rain saturates slopes.

Rivers swell.

Loose material becomes unstable.

Cloudbursts can add enormous amounts of water to small catchments within a short period.

And when a slope already contains fractured rock, sediment, snow or ice, heavy rainfall can increase instability.

But an important scientific caution is necessary:

Not every Himalayan flood is simply a “monsoon flood”, and not every glacier-related disaster can automatically be attributed to climate change.

For this specific event, ICIMOD has explicitly cautioned that it is too early to determine the role climate change played.

5. WHY THIS IS ALSO EARTHQUAKE COUNTRY

Here two different hazards overlap geographically.

The Himalaya exists because the Indian tectonic plate continues to converge with the Eurasian plate.

USGS estimates relative plate convergence around Nepal at roughly 40–50 millimetres per year, with a substantial component absorbed in uplift and deformation of the Himalayan mountain belt. The major plate-boundary structure beneath Nepal is the Main Himalayan Thrust.

Central Nepal experienced the catastrophic Mw 7.8 Gorkha earthquake in 2015, followed by a Mw 7.3 event. USGS research shows that earthquake shaking across this terrain can generate widespread landslides and avalanches.

EARTHQUAKE–FLOOD CONNECTION

Earthquake
↓

fractured or destabilised slope
↓

landslide / rock avalanche
↓

river blockage or debris entry
↓

possible flash flood

But this mechanism should not be confused with evidence about the current disaster.

There is presently no basis for saying an earthquake caused the 26 August flood.

ICIMOD noted unusual seismic signals but said no causal relationship had been established. USGS subsequently described the event as a glacial-collapse/debris-avalanche and flash-flood disaster.

A landslide or glacier collapse itself can generate a seismic signal.

That difference is critical.

WHY FLOOD AND EARTHQUAKE RISK OVERLAP HERE

It is therefore not that every flood is caused by an earthquake.

It is that the same young, steep, tectonically active mountains create conditions favourable to both hazards — and sometimes allow one hazard to trigger another.

A RIVER CORRIDOR ALSO CARRIES ROADS, POWER AND PEOPLE

There is another reason the Trishuli system matters.

Himalayan valleys are among the few places where infrastructure can be built through extremely rugged terrain.

Consequently, the same narrow corridor may contain:

• highways,
• bridges,
• hydropower plants,
• transmission infrastructure,
• markets,
• villages,
• tourism facilities,
• border trade installations.

IFC’s basin study has identified extensive hydropower development throughout the Trishuli basin and has previously warned that hydropower, slope instability, development pressures and other stresses need to be considered collectively rather than project by project.

When the river occupies almost the entire usable valley floor during an extreme event, everything built beside it can become part of the flood channel.

THE 26 AUGUST FLOOD: A DISASTER MOVING DOWN A MAP

The scale of the present tragedy is still being established, and casualty and missing-person figures have changed rapidly as rescuers reach previously inaccessible areas.

By 28th August – over 580 deaths across Nepal and China and nearly 2,500 people still missing – while authorities were simultaneously warning about additional risk from water accumulating behind debris blockages. Those numbers should be treated as a timestamped snapshot, not a final toll.

But the physical geography is already telling a larger story.

The disaster began high in one of Earth’s most difficult mountain environments.

The surge entered a narrow river.

That river became another river downstream.

More tributaries joined.

The gorge eventually opened into foothills and plains.

And places far removed from the initial collapse suddenly became part of the same emergency.

ONE FLOOD, DIFFERENT NAMES

BHOTE KOSHI / LHENDE

Upper Himalayan hazard zone

↓

TRISHULI

Fast mountain transmission corridor

↓

NARAYANI

Large downstream trunk river

↓

GANDAK

Indian plains

↓

GANGA

The terminology changes.

Gravity does not.

National borders change.

The watershed does not.

Administrative districts change.

The flood wave follows the river.

That is perhaps the most important way to read the geography of the Bhote Koshi, Trishuli and Narayani after the disaster of 26 August 2026.

A mountain flood is not simply an event at the place where the mountain fails.

It is a river-system event.

And in the Himalaya, the distance between a glacier, a gorge, a hydropower station, a highway, a village and a densely settled plain can be measured not only in kilometres —

but in how quickly the water gets there.