The worst storm you remember is not the worst one coming

What happened

Researchers at King's College London have published a warning about how we plan for weather disasters. It appeared in the journal Nature Sustainability on 10 September 2026. The lead author is Tom Matthews. He and his colleagues looked at deaths from every recorded tropical cyclone and every recorded extreme heat event between 2000 and 2025. Their argument is that the historical record is a poor guide to how bad things could get, because the record is short and because the climate that produced it no longer exists. Dr Matthews put it starkly: "History is almost guaranteed to make us underestimate future risk." He described recorded disasters as "the worst that have happened so far, in a short record, in a climate that no longer exists". The team says the what-if scenarios used in risk planning are not designed to find the worst possible outcomes, which means they are not designed to prevent them either.

Why this matters

The numbers behind the argument are striking, and they explain why averages mislead. Cyclone Nargis hit Myanmar in 2008 and killed around 140,000 people. That is more than twice the combined death toll of the other 675 deadly tropical cyclones recorded across the whole 25-year period. The 2003 European heatwave killed roughly 70,000 people, and nearly half of those deaths happened in the first two weeks of August alone. In other words, the deaths are not spread out evenly. They are concentrated in a handful of enormous events. So if you plan around the typical storm or the typical hot spell, you are planning for the wrong thing. This is not an abstract point for the UK. England had five heatwaves between May and August 2026, and the NHS has just recorded its busiest summer ever.

What the evidence actually says

This is a perspective paper, which is a particular kind of scientific publication. The authors analysed real data, but their main product is an argument about method rather than a new measurement. So it is not a forecast. Nobody is predicting a specific storm or a specific death toll. The three things they want planners to start doing are worth understanding. The first is horizon scanning, deliberately bringing in people who are not the usual experts, including insurers, frontline communities, researchers who study catastrophic risk, and even climate fiction writers, on the grounds that imagination is the point. The second is threshold mapping: working out the point at which impacts stop rising gradually and suddenly jump. A city may cope fine up to a certain temperature and then fail quickly above it. The third is looking at the physical limits of how hot it can actually get, rather than only at what has already happened. They also flag compound hazards, where two things land together. A severe heatwave immediately before a major storm, for example, when people are already exhausted and the power grid is already strained.

Qurexa perspective

We deliver medicines, and heat is the thing that quietly complicates that work. Most medicines are meant to be stored below 25°C, and some insulin and other products need a fridge. A hot week is a genuine problem for anyone whose medicine sits in a sunny porch or an unshaded delivery box. It is also the reason we would rather deliver early in the day during a heatwave, and why we would always rather you told us a delivery arrived warm than quietly used it anyway.

Practical advice

Have a plan for a very hot week, not just a warm day. Know which room in your home stays coolest and be willing to sleep there. Check your medicines. Read the storage line on the box. Keep them out of conservatories, parked cars and window sills. If something needs a fridge, know what you would do in a power cut. Know who is most at risk. Older people living alone, babies and very young children, people with heart, lung or kidney conditions, and anyone on water tablets, some blood pressure medicines or some mental health medicines. If that is a neighbour or a relative, make the call. Keep the basics in one place: bottled water, a torch, a battery radio, a phone power bank and a printed list of your medicines and doses. That list is far more useful in an emergency than people expect. Sign up for local flood warnings if you are in a flood-risk area, and take heat-health alerts seriously rather than treating them as nice weather news.

What to know

Researchers at King's College London, writing in Nature Sustainability on 10 September 2026, warn that the record of past disasters makes us underestimate future ones. Deaths are concentrated in rare, enormous events: Cyclone Nargis in 2008 killed more than twice as many people as the other 675 deadly cyclones of the last 25 years combined. They want planners to imagine worse, map the thresholds where things break, and study what is physically possible rather than only what has happened. For households, the useful version is simple: plan for a bad week, know where your medicines are stored, and know who nearby would struggle. Sources: Phys.org, "Future catastrophic weather could be 'far worse' than history suggests", 10 September 2026, https://phys.org/news/2026-09-future-catastrophic-weather-worse-history.html | Nature Sustainability, 2026, DOI 10.1038/s41893-026-01929-1, https://doi.org/10.1038/s41893-026-01929-1 | King's College London, Department of Geography, https://www.kcl.ac.uk/geography This article is for general information and does not replace advice from a doctor, pharmacist or other qualified healthcare professional.