Research

April 23, 2026 / 1 minute

AI predicts wildfire risk faster than current warning systems, new study finds

A wildfire forecasting system powered by artificial intelligence was around 30% better at identifying dangerous fire conditions in advance than the models currently used, according to new research. The Australian and New Zealand researchers’ study, published in the International Journal of Wildland Fire, trained a machine learning model on historical weather

Associate Professor Michelle Lupton and Professor Anthony White examine filters of wildfire smoke particles. Photo: QIMR Berghofer Medical Research Institute.

April 9, 2026 / 8 minutes

The smell of danger – understanding wildfire smoke exposure

By Anthony White At a laboratory in Australia, researchers are trying to develop rapid, accessible tests that can protect firefighters, military personnel, and the wider community from the escalating health burden of wildfire smoke and polluted air. The olfactory tissue, located high in the nasal cavity, directly connects the outside

March 22, 2026 / 2 minutes

When are wildfire evacuations most dangerous? Research finds leaving late risks lives

“Once you’re on the road, it may already be too late.” The somber warning was recently given by researchers who analyzed hundreds of self-recorded videos from people evacuating from wildfires. They were studying driving behaviour during emergency evacuations to learn why and when people choose to evacuate and understand how

March 3, 2026 / 1 minute

Synchronized wildfires will strain global fire resources: report

While the trend to more extreme fire weather days is well understood for many regions worldwide, a new study has identified the rising incidence of widespread concurrent extreme fire weather days across multiple global regions. The increased risk of concurrent fires poses significant social and ecological consequences and may complicate