From a glacier collapse in the Nepal-Tibet border, to wildfires across Europe, the past three months have shown that the climate crisis is no longer a forecast. It is a present tense reality

On the morning of 26 August, Rajendra Dawadi was inside his school in Nepal’s Trishuli Valley when warnings began arriving. Floodwaters were moving rapidly through the valley. A school accountant raised the alarm. A colleague called. A parent arrived to collect a child. The information was incomplete and the danger had not been formally confirmed. Dawadi had minutes to decide whether to evacuate the school or not.

He ordered the evacuation. According to reports in The Guardian and AP News, more than 900 students were moved to higher ground before a torrent of water, mud and debris swept through the area and swept way the school. “I decided I should not delay any more,” Dawadi told reporters. His decision saved the lives of numerous students. [1]

Elsewhere, settlements disappeared. Roads, bridges, hydropower infrastructure and the Gyirong border crossing between Nepal and Tibet were damaged or destroyed. The flood travelled through river systems, carrying rock, ice and debris downstream. The death toll later rose above 1,300 in published reports, while thousands remained missing. Figures changed rapidly and still require reconciliation with Nepal’s disaster authority.

The precise geological sequence remains under investigation.The United States Geological Survey described the event as a likely debris of avalanche and flash flood involving a glacier in the Langtang area. Scientists have not yet established exactly how much human driven climate change contributed to the collapse itself. That caution is important. It does not make the disaster less relevant to the climate crisis. Warming is changing the conditions around glaciers, permafrost, rainfall, slopes and river systems, while development places more people and infrastructure in harm’s way. [2] For people caught in the Trishuli Valley, however, the distinction between a geological trigger and a climate risk was not an academic one. They experienced the consequences as a wall of water.

The climate crisis is no longer a forecast. It is a present tense reality, measured in deaths, destroyed homes, disrupted economies, displaced families and ecosystems pushed beyond their limits.

The Nepal disaster also carried a large economic price. A government led assessment reported around US$2.7 billion in physical damage and economic losses, with recovery and reconstruction needs estimated at US$4.7 billion. The assessment included damage to roads, bridges, homes, schools, health facilities, agriculture and energy infrastructure. These figures are provisional and subject to revision. [3]

A summer of fire, water and exposure

Nepal was not an isolated shock. Across Europe, prolonged heat and drought created dangerous fire conditions from the Iberian Peninsula to the Balkans. The European Forest Fire Information System reported that 668,035 hectares had been affected by fires by 20th September. That is a measure of burned area, not a complete account of deaths or economic loss. It nevertheless captures the scale of an ecological emergency. [4]

In Greece, the emergency response became part of the tragedy. Two firefighting helicopters collided while operating near Psatha, west of Athens, as crews battled a wind driven wildfire. Two crew members died. The accident showed the conditions under which emergency personnel now work: high temperatures, shifting winds, poor visibility, exhausted crews and several fires competing for the same resources and infrastructure. [5]

The fires also damaged homes, forests, farms, tourism infrastructure and local economies. Fire prevention cannot mean only dousing the flames. It requires landscape management, safer construction, controlled burning where ecologically appropriate, protection of rural communities and investment in local fire services.

Europe’s wildfire risk may intensify as warming produces hotter, drier and windier conditions. A recent study (2026) in Global Change Biology projects that burned area could rise by 39 per cent under a lower emissions pathway and 192 per cent under a higher pathway by century’s end, showing why prevention matters [7].

Across September, a wider succession of climate- and weather-related disasters unfolded alongside the headline events. Flooding, landslides and intense rainfall affected Japan, Mexico, Myanmar, India and parts of southern Europe, while South Asia faced widespread monsoon flooding, landslides and displacement. East Asia experienced a combination of heat, heavy rain and tropical cyclones, and wildfires and smoke affected Canada and the United States. Large fires Satellite monitoring also recorded large fires in Mozambique and Zambia during the week. GDACS estimated that fires detected in Mozambique had affected almost 15,000 hectares and fires in Zambia almost 11,700 hectares. [8] These events underscore the geographically dispersed nature of the hazards. These events were not all individually attributable to climate change. They were, however, part of a global pattern in which hazards meet exposed settlements, fragile infrastructure and unequal access to protection. [6]

The disaster is never only the weather

The immediate cause of a disaster may be a fire, a flood, a landslide or a storm. A warming atmosphere can hold more moisture, increasing the potential for intense rainfall. Higher temperatures dry soils and vegetation, lengthen fire seasons and increase water demand. Glacier retreat, thawing permafrost and changes in snow cover can destabilise mountain slopes. Warmer oceans can add energy to tropical storms. None of these processes determines the outcome alone. The simple rule is this: weather describes what happened in the atmosphere; climate describes how the underlying conditions and probabilities are changing; attribution assesses how much human caused warming altered a particular event; and vulnerability helps explain why a hazard became a disaster.The final level of destruction depends on land use, infrastructure, warnings, poverty, governance and preparedness.

The World Meteorological Organization’s assessment of July and early August described a global climate marked by heat, drought, wildfire and flooding. July 2026 was among the warmest Julys’ on record, while extreme heat and very dry conditions affected ecosystems, agriculture, rivers, transport and energy systems (See Table: Climate and weather extremes unfold across the globe). The table below is a concise update. Figures are provisional, and “climate reading” distinguishes confirmed attribution from broader climate risk relevance.

Event and date	Reported impact	Loss or scale
Nepal and Tibet, 26 August	More than 1,300 reported dead; thousands missing; 900 students evacuated from one school.	US$2.7bn reported damage and losses; US$4.7bn recovery and reconstruction needs.
Europe, July to 9 September	Wildfires, evacuations, smoke and ecological damage across several countries.	658,520 hectares affected by 9 September.
Indonesia, 13 September	Ferry overturned in the Java Sea; six dead and 129 missing in initial reports.	More than 100 rescued; economic losses not yet consolidated.
Japan, 7 to 11 September	Flooding and landslides in Ibaraki; one death and about 3,320 displaced.	Economic losses not yet consolidated.
Mexico, 9 September	Storm and flooding in Michoacán; eight deaths.	About 5,000 people lost electricity.
Myanmar and India, early to mid September	Flooding in Yangon and floods and landslides in Mizoram.	GDACS recorded 15 deaths and 10,233 displaced in Myanmar; two deaths in Mizoram.
Southern Europe and Africa, 9 to 14 September	Flooding in Calabria and the Balearic Islands; large fires in Mozambique and Zambia.	One death and 858 displaced in Calabria; 17,000 without electricity in the Balearics; satellite detected large fire areas in Africa.

We had warnings. We treat them as paperwork

Dawadi’s decision in Nepal shows what early warning looks like when somebody acts. It also reveals the weakness of systems that leave life saving decisions to individual courage and informal phone calls. A school principal should not have to make a split second judgement while a glacier related flood travels down a valley. Warnings should be official, rapid, trusted and linked to rehearsed evacuation plans.

Forest fires reveal a related gap. Governments have invested heavily in emergency response, but prevention has struggled to keep pace with the changing conditions. Firefighting aircraft and helicopters cannot compensate for poorly managed landscapes, declining rural populations, accumulated vegetation and inadequate building standards.

Healthy landscapes require ecological management, controlled burning where appropriate, restored mosaics of land use, protection of biodiversity and the involvement of communities that live with fire risk. The aim should be to reduce the speed and intensity of fires while protecting people, ecosystems and livelihoods.

These are not merely environmental measures. They are economic protection. A wetland is a form of flood infrastructure. A functioning forest can reduce erosion. A reliable river gauge can protect a bridge, a school and a transport corridor. A safe evacuation route can prevent a disaster from becoming a mass casualty event.

The United Nations Early Warnings for All initiative aims to protect everyone from hazardous weather, water and climate events by the end of 2027. The system requires more than a message on a phone. It requires risk knowledge, monitoring, communication and the capacity to respond. [10]

The technology already exists in many places. Satellites can detect changes in glaciers and fire activity. Weather models can identify dangerous rainfall. Mobile networks can transmit alerts. Local knowledge can identify routes, safe buildings and vulnerable households. What remains weak is the connection between information and authority. Who issues the warning? Who closes a road? Who evacuates a school? Who pays for transport? Who checks on people without mobile phones?

These are governance questions, not technological mysteries.

Where do we go from here?

Preparedness must become a permanent function of government rather than an emergency reaction. That means investing in early warning systems and ensuring warnings cross borders, as disasters are not bound by human created borders. Nepal, China and India need mechanisms for sharing information about glacier hazards, temporary lakes, river levels and downstream evacuation risks. Similar arrangements are needed for shared rivers, cyclones, wildfires and heatwaves elsewhere.

Governments need to redesign development around present risk. Schools, hospitals, bridges, power stations, roads and water systems should not be rebuilt according to historical climate assumptions that no longer hold. Risk assessments must account for compound events, such as heat combined with drought or heavy rain falling on already saturated ground.

Adaptation finance should move from the margins of climate policy into the core of public investment. Funding is needed before disaster strikes, not only after homes have been destroyed. International finance should reach communities that face the greatest risks, including those least responsible for the emissions driving global warming.

Personnel and institutional capacity must follow the risk. Meteorologists, hydrologists, geologists, public health specialists, firefighters, engineers, social workers, planners and local officials, all need a role in climate risk management. Public agencies may need to redeploy staff and budgets from lower priority activities to early warning, prevention, emergency services and recovery.

Ecosystems must be treated as protective infrastructure. Forests, wetlands, mangroves, rivers and coastal systems are not scenery surrounding the economy. They are part of a sustainable planet. Destroying them creates new risks that are later paid for through public disaster budgets.

The world already has the Paris Agreement, the Sendai Framework for Disaster Risk Reduction and the Early Warnings for All initiative. The problem is not the complete absence of agreements. It is that many commitments remain weakly enforced, underfunded or disconnected from everyday decisions about land, infrastructure and public budgets. [10] [11]

A binding protocol for climate related disasters could require rapid data sharing, common warning standards, mutual aid agreements, satellite access, cross border rescue arrangements and predictable finance for recovery. A single global treaty covering every hazard may not be politically realistic. A set of binding regional agreements could be more effective. The essential point is that transboundary climate risks cannot be managed through voluntary goodwill alone.

The climate emergency is not waiting for the next COP. It is already testing whether governments can protect people from risks they understand. The answer will be measured in warnings that reach people, schools that stand, bridges that hold, forests that survive and lives that are not lost.

[1] BBC, “Head teacher evacuated 900 children minutes before Nepal floods hit” (30 August 2026). https://www.bbc.com/news/articles/c5ymvdm5520o

[2] United States Geological Survey, “2026 Nepal Debris Avalanche and Flash Flood” (2026). https://www.usgs.gov/programs/landslide-hazards/science/2026-nepal-debris-avalanche-and-flash-flood

[3] Business Standard, “Nepal estimates $4.7 bn recovery and reconstruction need after flash floods” (13 September 2026). https://www.business-standard.com/india-news/nepal-estimates-4-7-bn-recovery-and-reconstruction-need-after-flash-floods-126091300435_1.html Secondary report of a government led assessment. Verify against the National Planning Commission report before publication.

[4] European Commission Joint Research Centre, “Current wildfire situation in Europe” (updated 13 September 2026). https://joint-research-centre.ec.europa.eu/scientific-activities/natural-and-man-made-hazards/forest-fires/current-wildfire-situation-europe_en Data reported as of 9 September 2026.

[5] Reuters, “Two dead in firefighting helicopter crash in Greece” (2 August 2026). https://www.reuters.com/world/europe/two-firefighting-helicopters-collide-west-athens-2026-08-02/

[6] Global Disaster Alert and Coordination System, event records for Japan, Mexico, Myanmar, India, Italy and Spain (10 to 14 September 2026). https://gdacs.org/ GDACS figures are indicative and should be checked against national authorities.

[7] Billing, M., Forrest, M., Bowring, S. P. K., Oberhagemann, L., Neidermeier, A. N., Hetzer, J., von Bloh, W., Müller, C., Rolinski, S., Hickler, T., & Thonicke, K. (2026). Future projections of burned area in Europe highlight the importance of human action. Global Change Biology, 32(8), e71043. https://doi.org/10.1111/gcb.71043

[8] GDACS, forest fire event records for Mozambique and Zambia (updated 13 September 2026). https://gdacs.org/ Satellite based estimates of burned area and exposed population.

[9] World Weather Attribution, “Flood impacts in Assam driven by high exposure and structural vulnerability” (2026). https://www.worldweatherattribution.org/flood-impacts-in-assam-driven-by-high-exposure-and-structural-vulnerability-amid-uncertain-rainfall-trends/

[10] United Nations, “Early Warnings for All” (date not stated on page). https://www.un.org/en/climatechange/early-warnings-for-all Verify currency before publication.

[11] United Nations Office for Disaster Risk Reduction, “Climate action and disaster risk reduction” (date not stated on page). https://www.undrr.org/implementing-sendai-framework/sendai-framework-action/climate-action-and-disaster-risk-reduction Verify currency before publication.