Himalayan Glacier Timeline 1985–2026: GLOFs, Nepal Floods and the Rising Climate Risk
Himalayan glacier timeline from the 1985 Dig Tsho GLOF to Kedarnath, Chamoli, Sikkim and the 2026 Nepal-China flood, plus the rising climate risk.
Latest Story
High in the Himalayas, a lake can look peaceful for years. Then a chunk of ice, rock or moraine can fail, releasing a wall of water, mud and debris through valleys where people live, trade, build roads and generate electricity. The 2026 Nepal–China border catastrophe showed how fast a glacier-linked disaster can erase infrastructure. But the warning signs go back decades — to a 1985 flood in Nepal that changed how scientists understood Himalayan glacier risk. This Himalayan glacier timeline traces that arc, from Dig Tsho in 1985 to the Gyirong border collapse of 2026, and explains — event by event — what actually caused each disaster.
🧠 What is a GLOF?
A glacial lake outburst flood (GLOF) happens when water stored in a glacial lake suddenly escapes — often after an ice avalanche, moraine failure, landslide, heavy rain or earthquake-triggered instability. The released water can become a fast-moving flood carrying rock, ice and debris downstream. Not every Himalayan flood is a GLOF: some are rock–ice avalanches, extreme-rainfall floods, landslides or debris flows that never involve a lake at all.
Himalayan glacier disasters: key questions
Four decades of Himalayan glacier risk, in nine points
- The 1985 Dig Tsho disaster in Nepal is a landmark Himalayan GLOF event and one of the most studied in the range.
- Not every Himalayan flood is a GLOF. Some are rock–and–ice avalanches, some are extreme-rainfall floods, some are landslides or debris flows.
- The 2013 Kedarnath disaster was a compound event — extreme, well-above-normal rainfall combined with the breach of the moraine-dammed Chorabari Lake — not a simple lake-only GLOF.
- The 2021 Chamoli disaster in India was linked to a massive rock–and–ice avalanche, not a classic lake outburst.
- The 2023 Sikkim disaster involved the South Lhonak glacial lake and became a major warning for hydropower risk in glacier-fed valleys.
- The 2026 Nepal–China disaster showed how glacier collapse and debris flows can destroy border infrastructure and hydropower assets.
- Climate warming increases glacier retreat and can expand dangerous glacial lakes, but each disaster still needs event-specific attribution before it is blamed on climate change alone.
- The biggest future risks sit downstream: towns, hydropower plants, roads, bridges, border crossings and pilgrimage routes built deeper into unstable mountain terrain.
- Monitoring and mapping have expanded steadily since the 1990s, but warning time for a fast-moving collapse can still be minutes, not hours.
Not every glacier disaster is a GLOF
Five mechanism labels this page uses carefully — mixing them up is the single most common error in Himalayan flood coverage.
GLOF — glacial lake outburst flood
A glacial lake bursts or overtops, usually when a moraine or ice dam fails after an avalanche, rockfall, ice collapse, heavy rain or lake-level rise. Water stored behind the natural dam is released suddenly. Example: South Lhonak, Sikkim, 2023.
Rock–and–ice avalanche
A mountain slope, glacier ice or rock mass collapses and accelerates downslope, with no lake required. It can fragment, entrain sediment and water, and trigger a flood if it lands in a river or a lake. Example: Chamoli, 2021.
Debris flow
Floodwater mixes with mud, rock and sediment into a dense, fast-moving mass — sometimes described as “liquid concrete.” The high sediment load is what makes it so destructive to bridges, roads and buildings.
Landslide-dam flood
A landslide or avalanche blocks a river, ponding water behind an unconsolidated natural dam. These dams are unstable and often fail within hours, releasing a sudden downstream surge. Example: Nepal–China border, August 2026.
Extreme-rainfall flood
Heavy rain overwhelms rivers and drainage across already-steep, saturated terrain. In the high Himalayas this can combine with glacial-lake instability in a single compound disaster. Example: Kedarnath, 2013.
Mislabelling has consequences
Calling every flood a “GLOF” hides the real lesson of each disaster. Chamoli needed better slope monitoring, not lake monitoring. Kedarnath needed rainfall and land-use planning as much as lake surveillance. Precise labels drive the right fix.
How a warming mountain becomes a downstream disaster
One possible cascading pathway — not every Himalayan flood follows every step.
A glacier disaster is not always one simple event. It can begin with warming, but the final trigger may be an avalanche, rainfall, a slope collapse or a natural dam failure.
In the Himalayas, the disaster often begins where nobody lives and ends where everyone does. Before judging the risk to any specific valley, five questions matter more than the glacier itself:
- Is there a glacial lake or a steep glacier valley upstream?
- Are there villages, roads, bridges or hydropower plants downstream?
- Is there an early-warning system in place?
- Are evacuation routes marked and known to residents?
- Is real-time monitoring shared across borders, where the catchment crosses one?
Himalayan glacier timeline: 1985–2026
Newest first. Each entry is labelled by its verified mechanism, not assumed by default.
Nepal–China border: a glacier collapse devastates the Gyirong crossing
On 26 August 2026, a large section of glacier and rock detached on the Tibetan side of the Nepal–China border, fell roughly 1,200 metres, and became an ice–rock avalanche that temporarily dammed the Lhende Khola. When the natural dam failed, a debris-laden flood surged down the Bhote Koshi and Trishuli, devastating the Rasuwagadhi/Gyirong border crossing and hydropower infrastructure. The USGS initially logged a magnitude-4.4 earthquake, then revised its analysis: no tectonic earthquake occurred, and the magnitude-5.2 seismic signal was generated by the collapse itself.
Early warning and glacial-lake mapping intensify
Two verified GLOFs bracket this period. In August 2024, a rock avalanche struck a glacial lake above Thame in Nepal’s Khumbu region, triggering a two-lake cascade that damaged 14 buildings including a school and health post. In July 2025, rapid growth and sudden drainage of a supraglacial lake on the Purepu Glacier sent a transboundary flood down the same Rasuwagadhi corridor that would flood again in 2026, destroying the Nepal–China Friendship Bridge.
2023
South Lhonak, Sikkim: a true glacial lake outburst flood
On the night of 3–4 October 2023, the moraine-dammed South Lhonak Lake released a catastrophic flood. Peer-reviewed research (published in Landslides) found that an unstable lateral moraine mass — up to 14.7 million m³ of frozen material — collapsed into the lake, generating a roughly 20 metre wave that breached the frontal moraine and released about 50 million m³ of water.
2021
Chamoli: the disaster first mistaken for a GLOF
About 27 million m³ of rock and glacier ice detached from the steep north face of Ronti Peak and became an extraordinarily mobile debris flow that raced down the Ronti Gad, Rishiganga and Dhauliganga valleys, scouring valley walls up to 220 m above the riverbed. Early reports called it a GLOF; a 2021 study in Science established a rock-and-ice avalanche origin with no lake involved at all.
2013
Kedarnath: a compound rainfall-and-lake disaster, not a simple GLOF
Rainfall roughly 3–4 times heavier than normal — one monitoring station recorded 315 mm in 24 hours — combined with rapid snowmelt to raise water levels in the moraine-dammed Chorabari Lake above Kedarnath. The lake breached during the extreme rainfall event, adding a surge of water and debris to already-devastating flooding on the Mandakini River. Both drivers mattered: this was extreme rainfall plus a lake breach, not a lake failure alone.
2000s
Glacial-lake monitoring expands across the region
Following Dig Tsho, scientists and regional institutions began mapping dangerous Himalayan glacial lakes more systematically. Glacier and glacial-lake inventories compiled through this period for Nepal, Bhutan and parts of the Tibetan plateau (drawing on ICIMOD-led and government survey work) became the baseline for identifying which lakes across the region posed the greatest outburst risk.
1985
Dig Tsho: the landmark GLOF that started the science
An ice avalanche struck the moraine-dammed Dig Tsho lake in the Langmoche valley. The end moraine dam collapsed, and the lake drained downstream over roughly four hours. The resulting flood destroyed the nearly-completed Namche Small Hydroelectric Project about 11 km downstream, took out fourteen bridges along the Bhote Koshi and Dudh Koshi corridor, and caused damage as far as 50–60 km downstream.
Seven events, seven mechanisms
A single table for the search engines and the AI answer engines: year, location, mechanism, trigger and impact.
| Year | Location | Event type | Trigger / mechanism | Impact |
|---|---|---|---|---|
| 1985 | Dig Tsho, Nepal | GLOF | Ice avalanche into lake → moraine dam breach | Namche hydel project destroyed; 14 bridges lost; 3 dead |
| 2013 | Kedarnath, India | Compound: extreme rainfall + lake breach | ~3–4× normal rainfall + Chorabari Lake breach | ~5,700+ officially dead/missing |
| 2021 | Chamoli, India | Rock–ice avalanche | Slope collapse at Ronti Peak, no lake involved | 200+ dead; Rishiganga & Tapovan hydropower wrecked |
| 2023 | South Lhonak, Sikkim | GLOF | Moraine mass collapse into lake → wave → dam breach | Teesta III dam destroyed; ~50M m³ released |
| 2024 | Thame, Nepal | GLOF (cascade) | Rock avalanche into lake → wave → two breaches | 14 buildings destroyed in Thame village |
| 2025 | Rasuwa / Purepu Glacier | GLOF (supraglacial) | Rapid supraglacial-lake growth & sudden drainage | Friendship Bridge destroyed; 9+ dead |
| 2026 | Nepal–China border (Gyirong) | Glacier collapse / debris flow | Glacier & rock collapse → river blockage → dam-break flood | Border crossing devastated; hydropower & roads lost |
The Himalayan risk map: who sits downstream
Glaciers, rivers, borders, hydropower and dense populations are all connected in these mountains — a failure high up can become a disaster far below.
Afghanistan & Pakistan
Western Hindu Kush Himalaya glaciers feed the upper Indus system. Pakistan’s Karakoram and Hindu Kush ranges carry their own well-documented GLOF history, separate from the events on this page but part of the same broader HKH risk pattern.
India
Uttarakhand (Chamoli, Kedarnath) and Sikkim (South Lhonak) sit directly downstream of glacier-fed valleys carrying hydropower, pilgrimage routes and mountain highways — exactly the exposure pattern this page tracks.
Nepal
Home to Dig Tsho, Thame and the Nepal-side damage from Rasuwa and the 2026 border flood. Nepal depends heavily on hydropower and mountain roads built in the same corridors its glacial rivers drain through.
Bhutan
Included in the regional glacial-lake inventories built up since the 1990s. Bhutan’s own glacial lakes are monitored under the same ICIMOD-led Hindu Kush Himalaya assessment framework referenced throughout this page.
Tibet / China
The source side of both the 2025 Rasuwa flood and the 2026 Gyirong border collapse. Upstream glaciers and lakes here feed rivers that cross into Nepal within hours, making cross-border data-sharing essential.
Shared river basins
The Indus, Ganga and Brahmaputra systems — the major basins this page’s events feed into — connect all five regions above. A glacier failure in one country’s headwaters is a flood risk in another country’s downstream valley.
Why Himalayan hydropower sits in the blast zone
Hydropower projects are built in steep valleys because that is where water has energy. But those same valleys can funnel debris flows, GLOFs and landslide floods with almost no warning. The pattern repeats across every event on this timeline:
- Dig Tsho, 1985: destroyed the nearly-completed Namche Small Hydroelectric Project 11 km downstream — the first major recorded case of a Himalayan GLOF wiping out a hydropower asset.
- Chamoli, 2021: the Rishiganga plant was destroyed and the larger Tapovan-Vishnugad project badly damaged; most of the dead were construction workers on site.
- South Lhonak, 2023: the 1,200 MW Teesta III dam was destroyed, worsening the downstream surge.
- Rasuwa 2025 and Nepal–China 2026: the Rasuwagadhi project and other Bhote Koshi/Trishuli plants were damaged, and border infrastructure was cut.
Design must account for rare but high-impact mountain hazards, not just normal river flow. Damaged hydropower infrastructure can also amplify a cascade — a failed or overtopped structure adds water and sediment to the flood, and its loss compounds the economic damage.
India, Nepal and China: a shared and unequal risk
The Himalayan glacier story is not only environmental. It is also geopolitical and infrastructural. Nepal depends heavily on hydropower and mountain roads that sit in the same corridors its glacial rivers drain through. India has major hydropower, pilgrimage and border infrastructure in Himalayan states from Uttarakhand to Sikkim. Tibet/China contains important upstream glaciers, lakes and river headwaters — both the 2025 Rasuwa flood and the 2026 Gyirong collapse originated on the Tibetan side of the border and did their worst damage in Nepal.
Cross-border disasters need cross-border monitoring and data-sharing. A glacier can collapse in one country and flood another within hours, which makes joint river monitoring, shared satellite data and coordinated early-warning protocols a practical necessity, not a diplomatic nicety — particularly along the Nepal–China border corridor that has now flooded twice in two consecutive years.
Climate risk: what warming does, and does not, explain
Climate change can increase glacier retreat, permafrost thaw, slope instability and glacial-lake expansion — but each disaster on this page still needs its own investigation. It is accurate to say warming raises the background risk across the Hindu Kush Himalaya; it is not accurate to say climate change directly caused any single event unless a specific attribution study says so.
| Event | Attribution status | What the evidence shows |
|---|---|---|
| Dig Tsho 1985 | Not formally attributed | Predates modern attribution science; treated as a foundational hazard case, not a climate-attribution study |
| Kedarnath 2013 | Possible contributing factor | Extreme rainfall was the dominant driver; a specific climate-change attribution for this rainfall event has not been established on this page’s sourcing |
| Chamoli 2021 | Possible contributing factor | Studies note warming and development raise Himalayan risk generally; the collapse itself was not formally attributed to climate change |
| South Lhonak 2023 | Strong published evidence | Peer-reviewed work found climate-driven glacier retreat and permafrost degradation amplified the hazard |
| Nepal–China 2026 | Not yet attributed | Investigation ongoing; rapid melt and high temperatures flagged as suspected contributors, not confirmed causes |
What ICIMOD reported (Mar 2026)
- Ice-loss rates across the Hindu Kush Himalaya have roughly doubled since 2000
- Loss accelerated after 2010, strongest in the eastern and central Himalaya
- Up to ~27 m of ice-thickness loss since 1975 across monitored glacier datasets
- Roughly 12% of glacier area lost between 1990 and 2020
What it does not mean
- Not “every Himalayan glacier melts twice as fast” — this is a regional aggregate
- Not proof that climate change caused any one specific collapse
- Not a prediction that a specific lake or slope will fail on a specific date
- Not the same metric as ICIMOD’s earlier mass-loss-rate assessments — keep the figures separate
Can scientists predict a glacier collapse?
Not reliably in every case. Monitoring can identify high-risk lakes and creeping slopes, but the exact timing of an outburst or collapse is very difficult to predict. Risk can be reduced through lake monitoring, controlled drainage, early-warning systems, hazard maps, safer infrastructure planning and evacuation drills — not through prediction alone.
Satellites — optical imagery, synthetic aperture radar and InSAR — let analysts map lake growth and measure slow slope deformation, producing before-and-after images of a collapse scar within days. This is powerful for understanding an event after it happens and tracking a building hazard over months or years. It is not yet real-time collapse detection, and a lake not on any “dangerous lakes” list can still fail, as Thame showed in 2024.
Explore More Timelines
People also ask
Frequently asked questions
Nepal–Tibet Flood Tracker
The living tracker for Chamoli, Sikkim, Thame, Rasuwa and the 2026 border flood.
Explore →
Himalayan Earthquake Risk
Faults, past quakes and the Central Seismic Gap — without predicting the next one.
Explore →
Himalayan Hydropower Risk
Why dams, tunnels and power projects face rising geological and glacier-flood exposure.
Explore →
Glaciers to 2050
What continued ice loss could mean for rivers, floods and hydropower through mid-century.
Explore →
Major Earthquakes Worldwide
See how Himalayan quakes compare to Japan, Turkey, Haiti and other global disasters.
Explore →
Related timelines on AiTimeline
✍️ Editorial note & sources
This is an editorial, AI-assisted explainer compiled from public sources. It is not disaster-response guidance; for safety information follow your national disaster-management authority.
How we classify events: each disaster is labelled GLOF, rock–ice avalanche, extreme-rainfall flood, landslide-dammed flood or a named compound of these, based on the best available post-event or peer-reviewed science.
Primary sources: US Geological Survey (USGS); International Centre for Integrated Mountain Development (ICIMOD), including the HKH Glacier Outlook 2026; peer-reviewed studies in Science (Chamoli 2021) and Landslides (South Lhonak 2023); the World Bank’s assessment of glacial lakes and GLOFs in Nepal; India’s National Disaster Management Authority; Nepal’s National Disaster Risk Reduction and Management Authority. Reporting sources: Reuters, Associated Press and other agencies for current-event detail.
Sources & further reading
Every dated entry above was checked against these references. Last reviewed 6 September 2026.
- World Bank - Glacial Lakes and Glacial Lake Outburst Floods in Nepal (Dig Tsho 1985 case)
- Indian Academy of Sciences - A study on the heavy rainfall event around Kedarnath (J. Earth Syst. Sci.)
- Shugar et al. 2021 - A massive rock and ice avalanche caused the 2021 disaster at Chamoli (Science)
- A massive lateral moraine collapse triggered the 2023 South Lhonak Lake outburst flood (Landslides, 2024)
- ICIMOD - Hindu Kush Himalaya glaciers losing ice at double the rate since 2000 (March 2026)
- ICIMOD - GLOF from Thyanbo glacial lake sweeps away Thame Village (2024)
- Stimson Center - Transboundary glacial floods on the China-Nepal border (Rasuwa 2025)
- Al Jazeera - Nepal-Tibet floods: what happened, what caused them and who is missing (2026)