The Climate Models That Now Predict Individual Weather Events — and Why It Changes Everything

A new generation of climate models can now attribute specific weather events to climate change with unprecedented precision. Here is what this means for insurance, law, and policy.

For most of the history of climate science, the standard response to the question 'is this specific weather event caused by climate change?' was some version of 'we cannot make causal claims about individual events, only about patterns and frequencies.' This was scientifically accurate. It was also increasingly unsatisfying to policymakers, judges, insurers, and ordinary people trying to understand their relationship to a changing climate.

A new generation of attribution science — powered by the same ensemble modeling approaches that have produced significant improvements in seasonal weather forecasting — is changing this answer. The European Attribution Science Consortium, which brings together researchers from the UK Met Office, the Royal Netherlands Meteorological Institute, and climate groups at several other European universities, published new methodology guidelines in February 2026 that formalize the approach for quantifying what they call 'event attribution probabilities.'

The method works as follows: run the current climate model with observed greenhouse gas concentrations, generate many simulations, and measure how often the observed weather event (or something as extreme) appears. Then run the same model with counterfactual pre-industrial greenhouse gas concentrations. Compare the frequency of the event in the two simulations. The ratio gives the change in probability — the extent to which climate change made the event more or less likely.

For the extraordinary March 2026 European heat event, preliminary attribution calculations suggest that climate change made it between 3.5 and 5 times more likely than it would have been in a pre-industrial climate. The event was not impossible without climate change — but it would have been extremely rare rather than merely unprecedented.

The practical implications for insurance pricing and legal liability are significant and are already being tested in courts across Europe. If climate change demonstrably increased the probability of a specific disaster that caused quantifiable damage, who is liable for the increased component of risk that human emissions created?