By 31 August, the toll of the ensuing calamity was around 920 — 903 in Nepal and sixteen across the border in Tibet, with nearly 4,800 people still missing: some 4,250 in Nepal, close to 600 of them foreigners, and another 546 on the Chinese side. At least 275 Indian nationals were among the missing, with 108 already rescued; many were pilgrims caught on the Kailash-Manasarovar route. Close to 900 workers at a dozen hydropower projects were unaccounted for — perhaps 500 of them trapped in flooded tunnels — and around 700 megawatts of generating capacity was knocked out in a single morning. Whether the trigger was an ice-and-rock avalanche or the sudden drainage of a glacial lake high in Tibet is still being established.

What is not in doubt is that this was no ordinary flood.

Not An Act Of God

It is tempting to write off a disaster of this magnitude under providence and move on. That instinct is wrong. The proximate cause — a collapsing face, a bursting lake — is only the last link in a chain that we humans have been forging for decades.

The existential cause is warming. The Hindu Kush Himalaya, the “third pole” that feeds the rivers of two billion people, is heating nearly three times faster than the global average. On the current emissions path, these mountains could warm up by more than five degrees. Permafrost that once cemented the high faces is thawing; glaciers are retreating, losing ice in the last decade some 65% faster than in the decade before; and meltwater is pooling into thousands of unstable lakes behind fragile dams of rubble and ice. The region now holds upwards of 25,000 glacial lakes, dozens of them classified as dangerous in Nepal’s basins alone. 

Each such dangerous glacial lake is a loaded gun pointed downhill.

Onto this destabilising landscape we have bolted the apparatus of modern ambition — run-of-river dams, border markets, pilgrim highways, worker colonies — often in the narrowest, most exposed gorges, exactly where a debris flow has nowhere to go but through. The mountains are failing because the climate is changing; the death tolls are rising because we keep building in the fall line. 

Nepal’s tragedy is the collision of the two.

India’s mountain valleys now carry the same freight Nepal’s did on 26 August — pilgrim convoys on single-lane roads, labour colonies pressed against dam sites, hill towns spilling onto old landslide debris — all of it crammed into the few flat metres that a Himalayan river will one day come back to claim. When the Rishiganga fell at Chamoli, it was workers inside the Tapovan tunnel who drowned in the dark. When Teesta-III failed in Sikkim, an army encampment and the towns below it paid the price. We have built our own exposure with our own hands, permit by permit.

They are one recurring event, wearing different names, and each rehearsal has been costlier than the last. The uncomfortable truth is that India’s Himalayan frontier is being urbanised, dammed and monetised faster than it is being understood — and the bill is coming due in lives. Yet after every disaster the standard response is to grieve, compensate, and rebuild the same structure in the same doomed gorge, as though the mountain will make an exception the next time.

The Vulnerabilities We Have Engineered

Three failures turn a mountain hazard into a governance catastrophe, and all three are fixable.

We continue to plan and pour concrete in precisely the gorges the science tells us to vacate. As far back as 2014, an expert committee formed after the Kedarnath disaster recommended dropping 23 of 24 proposed Uttarakhand hydropower projects because they sat in the paraglacial danger zone. Most of that advice was quietly shelved. 

The second is warning. When South Lhonak and Chamoli struck, the people downstream had no siren, no cell-broadcast alert, no rehearsed evacuation — only the sound of the water. India has world-class space and sensing capability; almost none of it reaches the last mile of a mountain village at 8 in the morning or, still worse, at midnight. 

A warning that arrives after the flood is not a warning.

The third is water diplomacy. Much of the danger sits upstream, in Tibet, beyond our writ. India and China share no binding treaty on the Brahmaputra — only a soft “Expert-Level Mechanism” and seasonal, revocable courtesies on hydrological data, which Beijing has withheld before during periods of tension. Against that backdrop, China’s planned mega-dam on the Yarlung Tsangpo at the Great Bend — envisaged at 60 to 80 gigawatts, three times the Three Gorges — is being built in one of the most seismically violent zones on earth, the country that produced the magnitude 8.6 Assam-Tibet earthquake of 1950. A cascade failure there would not stop at the border. It would arrive in Arunachal and Assam as a wall of water we were never told was coming.

There is a bitter symmetry here that India too often forgets. 

On the Brahmaputra, we are the anxious downstream nation, pleading with China for data we cannot compel. But on the Kosi, we are the upstream-armoured power that Nepal accuses of the very same high-handedness — the 1954 Kosi Agreement is still resented in Kathmandu as unequal, and in August 2008, the Kosi breached its embankment at Kusaha, inside Nepal, and displaced close to three million people across Bihar.

A country that wants a “no-harm” norm honoured above it must be scrupulous about honouring one below it. Credibility on transboundary water is indivisible; we cannot demand of Beijing what we withhold from Kathmandu and Dhaka.

What India Must Do Now

A serious Indian response rests on a handful of structural fixes, each within the reach of policy rather than miracle.

Firstly, plan for chains, not floods. Our disaster machinery still treats a landslide, a dam breach, and a flood as separate hazards. In the mountains they are one cascading process, and our early-warning, evacuation, and infrastructure codes must be rewritten around that reality.

Secondly, instrument the high country and close the last mile. Every lake flagged as dangerous should carry sensors; every valley below it should have sirens and automatic cell broadcasts tied to rehearsed evacuation drills. This is cheap relative to a single lost dam.

Thirdly, audit and lower the dangerous lakes. Sikkim began siphoning South Lhonak only after it killed. The inventory of hazardous glacial lakes exists; the mandate and money to physically lower the worst of them does not. That must change, lake by lake, before the breach, not after.

Fourthly, stop building in the death zone. The 2014 committee’s advice should become binding: no new hydropower or township in the identified paraglacial hazard belt, and a hard second look at what is already there.

Fifthly, enact the zoning law we already wrote. The Model Flood Plain Zoning Bill has existed since 1975; in half a century only a handful of states have enacted it. A river’s floodplain is the river’s, and no amount of construction permission repeals that.

Then, turn upstream data into a right, not a favour. India must press, bilaterally and through a wider Himalayan compact with Nepal, Bhutan and Bangladesh, to convert Chinese hydrological courtesies into guaranteed, real-time, treaty-bound sharing — and to establish a “no-harm” norm around projects like the Yarlung Tsangpo dam that can kill in another country. Neither China nor India has ratified the 1997 UN Convention on international watercourses. The glaciers do not recognise the Line of Actual Control, and neither will the next flood.

Further, build the science to see it coming. The inventories of dangerous lakes, the hazard maps, the paraglacial zonation — these exist today as one-off studies and expert reports that gather dust between disasters. They must be consolidated into a permanent, funded, authoritative Himalayan hazard institution whose findings are binding on the ministries that grant construction and pilgrimage permissions, not advisory footnotes they are free to ignore.

Finally, move the money to the front of the disaster. We spend lavishly on relief and stingily on prevention. Every rupee shifted from post-flood compensation to pre-flood mitigation — sensors, zoning, the physical lowering of lakes, the resettlement of people out of the fall line — buys back lives at a fraction of the cost, and buys them before they are lost rather than mourning them afterwards.

The Reckoning

The Himalaya gave notice in Nepal. It has given the same notice, repeatedly, on our own soil — at Kedarnath, at Chamoli, at Lhonak — and each time we have grieved, compensated, and rebuilt in the same doomed gorge. A five-degree Himalaya is now a planning assumption, not a distant scenario. The mountains will keep falling; the only variable within our control is whether we are standing beneath them, unwarned, when they do. 

(The writer is a former civil servant. Views are personal.)