The threat is so great, in fact, that any official tally of people killed in a fire most likely is wildly low, given that it counts obvious victims, not those who later died after inhaling its far-flung haze. Los Angeles' catastrophic blazes in January, for instance, killed 30 people according to authorities, but more like 440 according to scientists, who determined excess deaths at the time were likely due to smoke.
Global warming intensifies wildfires and exacerbates greenhouse gas and pollutant emissions1. However, global projections remain incomplete, hindering effective policy interventions amid uncertain warming futures2. Here, we developed an interpretable machine learning framework to project global burned areas and wildfire emissions. This framework accounts for the impacts of future climate change on fire activity and quantifies associated premature deaths and radiative forcing from fire-induced particulate matter (PM2.5).