Why Nepal’s glacial lakes need Indian action

This article is authored by Brabim Karki, columnist and commentator, Nepal.

Nepal’s ambitious, six-year plan to drain four high-altitude glacial lakes—Thulagi, Lower Barun, Lumding Tsho, and Hongu-2, is born of sheer urgency. Following a catastrophic flood along the China-Nepal border that left more than 1,450 dead and thousands missing, Kathmandu is scrambling to mitigate future disasters. New Delhi cannot afford to view this as a purely domestic Nepali affair. All four targets sit on a 2020 ICIMOD-UNDP high-risk inventory within the Koshi and Gandaki basins, watersheds that feed the Koshi and Gandak rivers sweeping into Bihar and Uttar Pradesh. For India, providing immediate technical support, data-sharing, and early warning systems to Nepal are proactive flood defence for its own northern plains.

This photo is provided by the Nepal Army Budhi Gandaki river in north central Nepal is seen in spate, Sunday, Oct. 11, 2026. (Nepal Army via AP) (Nepal Army via AP)

The geography of transboundary disasters is unsparing. The catastrophic August deluge surged from the Lhende Khola into the Bhote Koshi, then wound through the Trishuli and Narayani rivers before crossing into India as the Gandak to join the Ganga. By the time such glacial bursts reach the lowlands, they lose their vertical momentum. The primary threat is rarely a sudden wall of water, but rather an acute, destabilizing volume surge delivered to rivers already swollen by the monsoon and hemmed in by fragile embankments. The precedent for disaster is well-established. Just two years ago, torrential rains forced Nepal to open every gate of the Koshi barrage, triggering flood alerts across 13 districts in Bihar. The 2008 Koshi embankment breach at Kusaha permanently altered the river's course.

With engineering design studies slated to begin in the fiscal year ending mid-July 2027, Nepal is drawing on past successes. It lowered Tsho Rolpa by three meters in 2000 for the first time and replicated the feat at Imja a decade later. Yet, engineering has its limits. The current project addresses only four of Nepal's 21 highly dangerous glacial lakes, leaving the exact valley that failed in August completely untouched. Where engineering fails, technology must step in. When the August collapse occurred, seismic instruments registered the shockwave as a magnitude 5.2 event. Linked to automated alerts, such data can buy downstream communities precious minutes, provided the monitoring systems are properly maintained.

However, technology requires infrastructure and upkeep. Automated monitoring stations, lake-level sensors, and satellite feeds are remarkably inexpensive when measured against the catastrophic cost of a breached levee. Facing a $5 billion recovery bill, Nepal cannot secure its mountains alone. India already possesses world-class satellite arrays, meteorological assets, and a Central Water Commission that tracks these very river systems. Sharing this existing data costs New Delhi virtually nothing. Furthermore, the broader geopolitical challenge includes Tibet, where many of the region's perilous lakes reside. Currently, critical alerts must traverse a friction-filled path from China to Nepal to India, stifled by a chronic lack of trilateral cooperation.

But the hydro-politics between New Delhi and Kathmandu carry historical baggage. The 1954 Koshi Agreement remains a point of friction in Nepal, and any unilateral Indian offer to plant equipment in Nepali mountains will be viewed with skepticism. New Delhi should route its cooperation through multilateral bodies like the UN Development Programme and ICIMOD, the Kathmandu-based mountain centre where India holds a seat. By ensuring data flows symmetrically and leaving station ownership with Nepal, India can build trust.

Early warning cooperation is a rare geopolitical arena where one nation's gain does not equal another's loss. During the August disaster, the river gauge at Syabrubesi went offline within minutes of the glacial collapse. A sensor's sudden silence is a loud distress signal, but only if someone downstream is listening. While lowering glacial lakes takes years, transmitting a digital warning takes seconds and costs a fraction of the infrastructure it protects. India's major rivers begin in foreign peaks; its domestic security is hostage to what happens in those mountains and how fast that information reaches Delhi and Patna. The August flood decimated a vital border trade route on its way south. The next deluge may not pause at the border.

(The views expressed are personal)

This article is authored by Brabim Karki, columnist and commentator, Nepal.