In general, extreme event attribution, also known as attribution science, evaluates relative contributions of multiple causes of an event, and assigns statistical confidence to that evaluation. Most often, the term is applied to the science of identifying and quantifying the role that human-caused climate change plays in the frequency, intensity, duration, and impacts of individual extreme weather events. Attribution science aims to determine the degree to which such events can be explained by or linked to human-caused global warming, and are not simply due to random climate variability or natural weather patterns.
Noted in “The secret weapon that could finally force climate action,” Thomas Lewton, NewScientist (10 January 2026, paywall):
But climate models can be put to another use if they are run in a slightly different way. The idea is to simulate counterfactual scenarios and compare them with how things really turned out. How would the world look if we had left a certain portion of fossil fuels in the ground, for example? Climate scientists have spent decades using this technique to figure out the consequences of carbon emissions, in a field that is broadly called attribution science.
