Vegetation State Key to Predicting Wildfire Severity in California, Study Finds

Vegetation Health Outranks Topography and Weather in Wildfire Severity

Urbanization, climate change, and decades of fire suppression have intensified wildfire risks in California’s densely populated wildland-urban interface. These combined factors make fires increasingly unpredictable and difficult to manage. The danger became starkly evident in January 2025, when massive wildfires tore through Los Angeles, consuming entire hillsides and canyons, devastating neighborhoods, and inflicting lasting damage on surrounding ecosystems.

California’s Fire-Prone Landscape

The Mediterranean climate region of California, which dominates the state’s coastline, is naturally fire-prone. Its dry conditions foster robust vegetation growth while simultaneously enabling rapid fire spread. As wildfires grow more intense, the need for advanced modeling and deeper understanding of their drivers becomes critical to predicting and mitigating risk.

Study Focuses on Three January 2025 Los Angeles Wildfires

In a 2026 study published in AGU Advances, researchers Ward-Baranyay et al. analyzed preburn conditions for three major wildfires that occurred in Los Angeles in January 2025. Their analysis included:

• Fuel characteristics
• Topography (elevation and slope)
• Wind speed

The team leveraged satellite observations from two NASA instruments:

• ECOSTRESS (Ecosystem Spaceborne Thermal Radiometer Experiment on Space Station)
• EMIT (Earth Surface Mineral Dust Source Investigation)

These instruments provided detailed data on vegetation health in the months leading up to the fires. The researchers then developed a random forest regression model to predict burn severity based on these preburn conditions. The results highlighted that prefire fuel conditions were the primary driver of the destructive wildfires’ immediate effects on wildlands.

Model Accuracy and Key Findings

The model achieved approximately 60% accuracy in predicting burn severity patterns. It performed best for the Palisades and Hughes fires but was less accurate for the Eaton Fire. The discrepancy may stem from the Eaton Fire’s more topographically variable terrain, suggesting that the model did not fully capture all burn severity drivers in that area.

Vegetation type also proved to be a strong performance indicator. Terrain covered with shrubs or scrub—California’s dominant vegetation type—yielded the most accurate predictions for burn severity. The model was less effective at predicting burn patterns in forested areas and other landscape types.

Fuel Conditions: The Dominant Driver of Wildfire Severity

According to the study, fuel conditions emerged as the most influential factor in determining burn severity, surpassing topography and weather. Key indicators included:

• Vegetation abundance
• Moisture levels
• Dryness
• Stress levels

The researchers concluded that monitoring these fuel conditions could serve as an effective strategy for assessing wildfire hazard in California and other fire-prone regions.

Citation

Owen, R. (2026), Want to predict wildfire severity? Look to the state of vegetation, Eos, 107, https://doi.org/10.1029/2026EO260130. Published on 4 May 2026.

Text © 2026. AGU. CC BY-NC-ND 3.0

Except where otherwise noted, images are subject to copyright. Any reuse without express permission from the copyright owner is prohibited.