Photo by Sergey Pesterev on Unsplash
Key points
-
Infrastructure planning, disaster preparedness, and insurance rely on outdated historical benchmarks that cannot match rapidly intensifying, less predictable climate risks, experts have warned.
-
The Hindu Kush Himalayas face compound hazards including glacier retreat, permafrost degradation, extreme rainfall, and unstable geology. These can trigger hard-to-detect failures and overwhelm infrastructure built for past climate averages, as seen during the 26 August flash floods along the Nepal-Tibet border.
-
Experts call for holistic environmental assessments, tougher climate benchmarks in tenders, and mainstreaming climate considerations into public budgets.
-
With insurance increasingly inadequate, they propose regional disaster pools, expanded data sharing and monitoring, and embedding climate risk into procurement, land-use, and fiscal planning while building fallback mechanisms to absorb unavoidable losses.
Approaches to infrastructure planning, disaster preparedness and insurance are “outdated” and unable to respond to increasingly complex physical risks, experts have warned.
“We base these plans on past benchmarks. And these benchmarks are becoming outdated very rapidly,” said climate-change economist Ulka Kelker, speaking at the webinar organised by research site Carbon Copy.
The problem is particularly acute in the Hindu Kush Himalayas (HKH), a region which provides water for 2 billion people. Glacier retreat, permafrost degradation, extreme rainfall and unstable geology can interact to produce cascading disasters, as demonstrated by the 26 August flash floods along the Nepal-Tibet border. But the systemic financial challenges extend far beyond the region, said Kelkar, who is executive director of climate, economics and finance at World Resources Institute, India.
As climate risks intensify and become less predictable, governments everywhere face larger potential losses while also having to spend more upfront on adaptation.
Infrastructure under a changing climate
The disaster has laid bare the difficulty governments have of anticipating hazards that do not fit established models.
The event was not a conventional glacial-lake outburst flood, for which early-warning systems can be relatively effective, said Farooq Azam, a senior cryosphere specialist at the International Centre for Integrated Mountain Development (Icimod). Instead, he said, the initial failure occurred beneath the rocky surface, making it considerably harder to detect.
Initial estimates suggested as much as 200 million cubic metres of rock and ice may have fallen roughly 1,200 metres into the valley, which triggered a rapid rise in localised temperatures that in turn rapidly melted the ice and contributed to the resulting floods.
Glacier retreat may also be altering the risk profile further, he explained, with more rock surface exposed to solar radiation and warming at high altitude degrading permafrost beneath the surface.
Azam said it was therefore essential that environmental assessments consider not only the impact infrastructure has on the environment, but also how a changing environment could affect the infrastructure itself.
“When the government is putting some big hydropower project, is there any glaciologist in the panel who is looking for this danger?” he asked. “If there is some flood event, what would the impact of this be on this infrastructure?” That approach, he said, is “completely missing.”
Hydropower projects present additional challenges, compounding the disaster, as water stored in dams can add to the volume of a flood when these are breached, Azam added.
Infrastructure planning must therefore weigh up the risks more holistically, the panellists argued, warning against historical assumptions which often fall foul of losses that were not reflected in the original design or risk assessments.
Past benchmarks are “becoming outdated very rapidly”, Kelkar said, while infrastructure needs to withstand a much wider range of possible flood levels, rainfall and other physical conditions.
A project designed to withstand past climatic averages would likely struggle in light of a climate disaster considered implausible at the time of construction, but which may have become increasingly likely during its operational life.
Kelhar urged governments to include wider climate ranges into infrastructure tenders and require projects to withstand higher flood lines and more extreme rainfall. While upfront costs were likely to increase as a result, she argued it would reduce the potential for financial loss in the case of an extreme event.
Climate considerations should also be mainstreamed into government budgets rather than treated as a separate funding stream, she said.
“We are not helpless. We are not sitting ducks,” Kelkar said, referencing the global debate around the mobilisation of international financial assistance on behalf of Nepal. She said Third Pole countries – encompassing both the HKH and the Tibetan Plateau – have a range of domestic options they can draw from, including the adoption of cumulative environmental assessments and engaging in greater regional technical cooperation, as well as tougher climate benchmarks in tenders.
Nepal’s human and financial toll
The floods have had a devastating toll. The country could need as much as US$5bn, equivalent to nearly 10% of its economy, Nepal’s finance minister Swarnim Wagle said shortly after the disaster.
As of 6 September, 1,341 bodies had been recovered and around 5,000 people remained missing.
“A preliminary assessment conducted under the rapid damage and needs assessment (RDNA) indicates that approximately 7,570 houses affecting around 32,933 people, and 48 government offices have been affected. Similarly, 55 kilometers of roads, 37 motorable bridges, and 68 suspension bridges have been damaged. Likewise, approximately 1,806 hectares of agricultural land, 13 hydropower projects, one solar power plant and two transmission lines have been also damaged,” the Nepalese government said in a statement. 783MW of hydropower capacity has also been affected, it added.
A comprehensive post-disaster needs assessment is already under way to determine the full extent of loss and damage and the cost of reconstruction.
The government formally appealed for financial support to its Prime Minister’s Disaster Relief Fund on 27 August. Nepal’s foreign secretary has also said that, while international assistance has already been received for rescue and relief operations, “a significant level of support” will be required once reconstruction begins.
Panellists on the webinar said the tragedy had exposed an urgent need for climate-vulnerable countries to finance resilience before disasters occur, rather than repeatedly financing reconstruction afterwards.
When insurance is no longer enough
Insurance, which has traditionally been one way of transferring disaster risk, is no longer fit for purpose.
“Because of how widespread and how repeated these climate-change events are, insurance as a mechanism is also no protection,” Kelkar said.
She suggested the implementation of a “regional disaster pool”, combining different sources of capital, from government compensation to development-bank funding, from insurance and reinsurance on market mechanisms, to be leveraged by individual countries in the event of an overwhelming disaster.
Such a resource would be particularly helpful in a Himalayan context, where transboundary resources merge with transboundary risks. Data sharing agreements are currently restricted to the monsoon season, which should be extended year-round by regional governments, who should also prioritise regional collaboration and a regional insurance mechanism to help address these collective challenges, Kelkar added.
Improved data would also assist with assessing risks that conventional warning systems may miss, and improve the pricing of infrastructure and disaster risk.
High-resolution satellite imagery and machine-learning models could potentially detect slow movements preceding major rock and ice failures, while more high-altitude weather stations would improve rainfall and temperature forecasts, according to Azam.
Given the unpredictable nature of future risks, the panellists said governments and financial institutions must resist the temptation to delay action in favour of the perfect forecasting model.
Instead, policymakers should incorporate climate risk as standard into infrastructure procurement, land-use decisions and public budgets while developing fallback mechanisms to absorb losses when adaptation cannot prevent them entirely.
Climate risk is no longer just an environmental concern. Physical shocks will increasingly become economic and fiscal shocks, with consequences for businesses, banks, insurers and the wider financial system.
Kelkar said: “We need to recognise that climate change is bringing a lot of cascading and compounding risks.”
This page was last updated September 7, 2026
Written by
Soraya Kishtwari is a Contributing Editor at Green Central Banking. Based in London, she has more than 17 years’ experience reporting on and leading coverage at the intersection of politics, policy and the environment. A former political journalist, Soraya worked for The Times and PA Media before freelancing overseas. As Asia-Pacific Editor at Dialogue Earth, she led coverage of climate diplomacy, transboundary resource management and the green energy transition across Asia and Southeast Asia.