A new study by scientists at King's College London suggests that future catastrophic weather events could be far more severe than historical records indicate. Published in Nature Sustainability, the research highlights that current methods for assessing climate risks may not be prepared for the most extreme scenarios. Historical data on deadly storms and heat waves may not accurately reflect the scale of future disasters, which could have far greater impacts on human lives, infrastructure, and health. The study warns that existing risk models are not designed to account for the worst-case scenarios, leaving a gap in how societies prepare for such events. The findings come after a summer in Europe marked by thousands of excess deaths due to heat waves, and amid concerns about the potential effects of the El Niño weather pattern, which can bring more extreme weather globally. The research team analyzed deaths from tropical cyclones and extreme heat events between 2000 and 2025. They found that the majority of fatalities occurred during a small number of massive events, and that future events could be far more severe than current expectations suggest. One example cited in the study is Cyclone Nargis, which killed nearly 140,000 people in Myanmar in 2008. That single event caused more than twice as many deaths as the combined total of all other deadly tropical cyclones during the 2000–2025 period. Similarly, the 2003 European heat wave resulted in about 70,000 deaths, with nearly half of those occurring in just two weeks in August. The study acknowledges that while it is difficult to rule out future events causing millions of deaths, the researchers emphasize that such outcomes are not guaranteed. Dr. Tom Matthews, a senior lecturer in environmental geography at King's College London and lead author of the study, noted that historical records are likely underestimating future risks. The climate today is different from the one in which past extreme weather events occurred, and the records used to understand these events are based on a short and changing history. To better prepare for future extreme weather, the researchers have proposed three main priorities: conducting "horizon scans" that involve diverse experts to imagine future risks, understanding the thresholds at which impacts become drastically worse, and mapping the physical limits of weather in climates far hotter or colder than today's. The study calls for a broader scientific and societal effort to recognize and prepare for unknown catastrophic events, involving fields as varied as climate fiction, insurance, and existential risk studies.