There is no such thing as an easy mountain.
In 2022, I was choosing my first 8,000-metre peak. Manaslu was the obvious candidate, the peak climbers around the world call the gentle introduction to the giants. That autumn, Nepal issued a record 404 permits for it, and with support teams, somewhere between 700 and 1,000 people were on its slopes.
On 26 September, an avalanche between Camps 3 and 4 killed Anup Rai, a 34-year-old Nepali climber carrying oxygen up to Camp 4. The same day, Hilaree Nelson, whom her sponsor called the most prolific ski mountaineer of her generation, triggered a small slide while skiing from the summit and fell to her death. Six days later, another avalanche flattened tents at base camp. Even Ralf Dujmovits, one of the most experienced high-altitude climbers alive, turned back in the wind and snow.
In 2024, I chose Annapurna instead to start my 8000m journey, and the mountain climbers whispered about. Manaslu taught me that no mountain is easy. The mountain makes the decisions. All we control is how ready we are when it does.
This autumn, Manaslu repeated the lesson. Heavy snow buried the mountain through September. On the night of September 19, an avalanche struck Camp 3 and buried Dhruba Jyoti Guha, a German climber of Indian origin, and his guide Phursemba Sherpa inside their tent. Theirs were the third and fourth deaths of the season. Within days, most expeditions had abandoned the mountain.
Three weeks later, the valley below paid its own price. At about 4 am on October 11, a torrent of mud and debris tore down the Budhi Gandaki in Gorkha. There had been no heavy rain that weekend. Authorities believe a dry landslide in Chumnubri dammed the river and then gave way. By 8.40 am, the river at Arughat stood above eight metres, against a danger mark of five. Between 35 and 40 houses were swept away, along with bridges, power towers and the construction camp of the 341 MW Budhi Gandaki hydropower project. Herders sheltering in sheds near the source are among the missing, and more than a hundred trekkers were stranded at Jagat.
It was at least the valley’s third flood in 30 months. In April 2024, an ice avalanche off the Manaslu Glacier crashed into Birendra Tal and pushed its water into the river. On September 6 this year, the Therang Khola surged, taking homes, bridges and a newly completed small hydropower plant that local families depended on. Three floods, and three different triggers. There is good news in this story, and it deserves credit. Nepal’s flood forecasters sent SMS alerts downstream at 6.40 am, before the surge peaked at Arughat, and the disaster authority’s chief believes they kept the human toll low. Warnings work. But they work downstream. The people nearest the source, in the high valleys where the landslide fell, had no instrument watching over them, and they are the ones still missing.
This should concern us in India. Our glacier risk programme is built around 195 high-risk glacial lakes, and that list matters. Yet none of the three Manaslu floods began with a lake bursting. Each began with a chain: snow loads a slope, the slope fails, debris chokes a stream, and the stream breaks free. Chamoli in 2021 began the same way, with rock and ice falling from a ridge. The hazard is rarely the thing we mapped. It is the sequence we did not.
And the water does not stop at the border. The Budhi Gandaki joins the Trishuli, which feeds the Narayani, which enters Bihar as the Gandak. This is where climate finance stops being an abstraction. Nepal produces about 0.05 per cent of global emissions. It is still rebuilding from August’s glacier-collapse flood, which killed more than 1,400 people and will cost an estimated $4.77 billion to repair. Its own adaptation plan puts the cost of glacial-flood risk reduction and early warning at $1 billion by 2030; by last year it had received about $7 million in international climate finance for the purpose. Mountain regions receive only about 3 per cent of global climate finance, and less than 1 per cent of adaptation funding goes to glacier protection. India’s own national programme runs on roughly $20 million. Watching whole valleys, from ridge to riverbed, costs money that mountain governments do not have, and loans that deepen their debt are the wrong tool for a bill someone else’s emissions ran up.
Next month, negotiators gather in Antalya for COP31. Three things belong on the table: grant-based adaptation finance set aside for mountain and cryosphere hazards, monitoring that reaches the high valleys where cascades begin, and a shared India-Nepal data system, because a landslide in Gorkha is a flood warning for Bihar. India knows what patient investment can do. After the 1999 Odisha cyclone, two decades of steady funding made the state a model for moving people before the storm arrives. The mountains need the same patience, and the same money.
On Annapurna, I learned that ice can hold a person as surely as it can harm one. It literally protected my brain while severe frostbite took away my digits. The mountain still decides. Preparation decides who is left standing afterwards.Manaslu has warned us three times. I would rather not learn what the fourth warning costs.
(Anurag Maloo is the founder and CEO of The Voice of Glaciers Foundation (TVGF) and Orophile Ventures. He is a World Economic Forum Young Global Leader 2025. In April 2023, he survived 72 hours trapped in a glacial crevasse at 6,000 metres on Annapurna)