On New York University Week: Is there a wildfire that we’re not planning for?
Augustin Guibaud, assistant professor of mechanical and aerospace engineering, examines this.
Faculty Bio:
Dr. Augustin Guibaud is an Assistant Professor in the Department of Mechanical and Aerospace Engineering and a member of the Center for Urban Science and Progress. His research focuses on hazards through physics-based modeling and physics-informed machine learning (PIML), with a particular emphasis on fire safety challenges in complex environments. He leads the IgNYte Lab, which investigates urban air quality monitoring during large-scale haze events, fire dynamics in low buoyancy conditions for space exploration missions, heritage conservation through the modeling of fire-structure interactions, city-scale fire modeling in densely populated areas, and land management strategies for wildfire resilience. Dr. Guibaud’s work bridges practical applications with fundamental lab-scale analyses.
Dr. Guibaud is an active member of the Fire Safety in Space ESA International Topical Team, the Combustion Institute, and the Structure group of the French CNRS/Ministry of Culture’s Chantier Scientifique Notre Dame de Paris. He also serves as an Honorary Lecturer at University College London in the Department of Civil, Environmental, and Geomatic Engineering.
His contributions have been recognized with several prestigious awards, including the 2020 Prix de la Chancellerie and the Distinguished Paper Award at the 39th International Symposium on Combustion.
Transcript:
Land managers use sophisticated computer models to predict wildfire risk decades into the future. These models account for climate change—rising temperatures, shifting rainfall patterns. But our research in Portugal revealed they’re missing something critical.
Forests don’t stay frozen in time. Over the decades, vegetation itself evolves. Trees die, new species move in, entire ecosystems shift. Current models treat forests as static, essentially planning for forests that won’t exist by the time we need those predictions.
We tested what happens when you include these vegetation changes in wildfire models, and the results were striking. In one climate scenario, models without vegetation predicted a fifty-nine percent increase in burned area by 2060. But when we included how forests would actually adapt to those conditions, that increase dropped to just three percent.
Even more surprising, some higher-emission climate scenarios actually showed decreased fire risk in Portugal. Local fire risk doesn’t always track with global warming trends. The climate conditions and how vegetation responds can diverge significantly from global patterns. Fire-prone species like eucalyptus might thrive in some areas under certain conditions while declining in others.
This matters for practical decisions happening today. Planting fire-resistant trees sounds smart, but if those species won’t survive future conditions, you’ve wasted resources. Worse, if fire-prone species will thrive instead, you’ve locked in elevated risk for decades. Because forest ecosystems take about a century to fully restore, today’s planting decisions echo for generations.
Read More:
[NYU Tandon School of Engineering] - Wildfire Prevention Models Miss Key Factor: How Forests Will Change Over Decades
[ConnectSci] - On the importance of both climate and vegetation evolution when predicting long-term wildfire susceptibility
