HMN 2026: How Models of multi-ignition wildfires could predict catastrophic events

When lightning strikes: Models of multi-ignition wildfires could predict catastrophic events
Since 2012, the largest fires in California and the Arctic-boreal region began as multiple ignitions that later merged.(A) The August Complex fire in northern California in 2020 had 10 separate ignition points (denoted by turquoise dots). (B) A wildfire in Yakutia, Russia, in 2021 had 27 ignition points. The maps illustrate the fire starts from FEDS/ABFA fire tracking datasets along with the 12-hourly progression of these fires (in color), showing only fire perimeters up to the 99th percentile of their final area. See fig. S1 for a comparison of fire tracking fire starts with official governmental records for the August Complex fire and a fire in Canada. Both maps are on the same spatial and temporal scale. (C) The location of these two fires in California and Siberia. Credit: Science Advances (2026). DOI: 10.1126/sciadv.adx6477

Multi-ignition wildfires are not overly common. But when individual fires do converge, the consequences can be catastrophic. The largest fire on record in California, the 2020 August Complex fire, grew from the coalescence of 10 separate ignitions.

New research on multi-ignition fires

In a new study, published in Science Advances, researchers at Lawrence Livermore National Laboratory (LLNL), the University of California (UC), Irvine and collaborators examine multi-ignition fires, calculating their impact and modeling the mechanisms behind them by leveraging the Department of Energy’s flagship Energy Exascale Earth System Model (E3SM). The work shows that when flames combine, they are disproportionately destructive: They spread faster, last longer, generate stronger atmospheric events and strain firefighting resources.

In California, the study found that multi-ignition fires make up only 7% of the total number of fires, but they contribute to 31% of the burned area in the state.

“Multi-ignition fires have a disproportionate effect on the burned area,” said LLNL scientist and author Qi Tang. “Although they are quite rare, their influence is large compared to single-ignition fires.”

Tracking and modeling fire behavior

Researchers at UC Irvine used remote sensing data to track multi-ignition fires and capture crucial data on their behavior. From there, the team at LLNL applied a simulation framework that captures fire-triggered thunderstorms (pyrocumulonimbus) and their downstream effects. The model connects the dots at a kilometer scale to create a big picture understanding of how wildfires ignite, move, merge, interact with regional atmospheric dynamics and thermal dynamics and trigger extreme thunderstorms.

“Pyrocumulonimbus events often occur when there’s an enormous wildfire event, but not all wildfires can trigger them,” said Tang. “The distribution actually is very non-uniform around the world.”

California, Canada and Siberia are some of the most likely locations for these fire-triggered thunderstorms. The extra heat from the fire at the surface creates a powerful updraft that lifts hot air and moisture into the sky. There, it condenses and can turn into a powerful storm. The same phenomenon often occurs on hot summer afternoons.

Challenges for firefighters and future prevention

Pyrocumulonimbus creates a higher chance of lightning strikes, and, depending on wind and environmental conditions, the clouds don’t always coincide with the original fire location. This provides the perfect storm: an opportunity for a fire to spawn and later converge into a multi-ignition event.

Dealing with multiple fire fronts can be particularly difficult and dangerous for firefighters, who can get trapped if new flames spring up to surround them.

With their new modeling framework, LLNL scientists aim to predict—and possibly prevent—these events.

“We can help the community of firefighters know where the pyrocumulonimbus is more likely to occur, and that can lead to a prediction of where the fire triggers and the larger event would be,” said Tang. “We might be able to do something to avoid multi-ignition events. That is one of our objectives.”

More observational data, collected by a 2026 NASA field campaign, will further enable this modeling. The team at LLNL also aims to integrate their simulations with energy infrastructure planning.

“If we know there are fire events, we can simulate them and how that would influence the power grid,” said Tang. “The results could potentially enhance U.S. energy security.”

Publication details

Rebecca C. Scholten et al, Multi-ignition fire complexes drive extreme fire years and impacts, Science Advances (2026). DOI: 10.1126/sciadv.adx6477


The content is provided for information purposes only.