What question did this study set out to answer?

To improve predictions of wildfire size and occurrence by adequately addressing extreme fire events.

May 15, 2026Earth s FutureOpen Access

Accounting for Extremes in Modeling the Size and Likelihood of Large Fires in the United States

View Full Paper

Ask AI

Bookmark

Authors

Framework analyzes large fire characteristics in ecoregions, suggesting tailored predictive models.

To improve predictions of wildfire size and occurrence by adequately addressing extreme fire events.
Analyzed data from 30,331 large fire perimeters from 1984 to 2024 using MTBS data.
Fitted Pareto Type II, lognormal, and Weibull distributions to full data and exceedance-based tail samples.
Evaluated distribution performance across 105 ecoregions and 10 Geographic Area Coordination Centers.
Models focused on the body significantly misestimate exceedance probability by up to three orders of magnitude.
Lognormal and Pareto Type II outperform Weibull for body distribution, while Weibull fits tail best at GACC scale.
Optimal distribution models vary by region, highlighting the need for tailored approaches in fire management.

Asadian et al. (2026) studied this question.

View Full Paper

Ask AI

Bookmark