Skip to main content

New Data in Action: ECOSTRESS Data Offer Significant Potential for Wildfire Prediction and Analysis

Using machine learning models, researchers identify high ECOSTRESS Water Use Efficiency ratios as the most influential pre-fire factor.

Predicting the occurrence and spread of wildfires is a challenge in regions all across the globe, but innovative approaches to leveraging Earth observation data may change the landscape of wildfire prediction. In a new Data in Action story, researchers combined NASA's Ecosystem Spaceborne Thermal Radiometer Experiment on Space Station (ECOSTRESS) and wildfire event data to develop wildfire susceptibility models. By assessing pre-fire vegetation and drought conditions, these models can help predict the likelihood of wildfire occurrence and spread. In one use case, researchers using these models were able to predict 90% of wildfires a week in advance of the fires occurring. 

Read "ECOSTRESS Data Offer Significant Potential for Wildfire Prediction and Analysis" for more details about this advance.

Details

Last Updated

Sept. 25, 2025

Published

Sept. 25, 2025

Data Center/Project

Land Processes DAAC (LP DAAC)