Using Airborne LiDAR to Predict Fire Risk in Mediterranean Forests

Jenn Hoskins
6th July, 2024

Using Airborne LiDAR to Predict Fire Risk in Mediterranean Forests

The Apulia region in southern Italy provided the setting for this study, where 104 field plots were established to collect ground-truth data for validating airborne LiDAR's ability to estimate fine dead fuel loads for wildfire management.

Image adapted from: Lin et al. / CC BY (Source)

Key Findings

  • Researchers at the University of Bari used airborne LiDAR data to estimate fine dead fuel load in Mediterranean forests
  • The study developed models using LiDAR-derived metrics to predict different categories of fine dead fuel load
  • The models showed moderate accuracy, with the highest accuracy in estimating litter and the lowest in 1-hour fuel load
Mediterranean forests are increasingly threatened by wildfires, with fuel load playing a crucial role in fire dynamics and behaviors. Accurate fuel load determination contributes substantially to wildfire monitoring, management, and prevention. A recent study conducted by researchers at the University of Bari aimed to evaluate the effectiveness of airborne Light Detection and Ranging (LiDAR) data in estimating fine dead fuel load[1]. This study focused on developing models using LiDAR-derived metrics to predict various categories of fine dead fuel load. Wildfires have complex impacts on forests, including changes in vegetation, threats to biodiversity, and emissions of greenhouse gases like carbon dioxide, which exacerbate climate change[2]. The influence of wildfires on animal habitats is particularly noteworthy, as they can lead to significant changes in native environments. The extent of these alterations in species and habitats plays a crucial role in shaping forest ecology. Thus, understanding and managing fuel loads in forests is critical for mitigating these impacts. The study by the University of Bari integrated field data with airborne LiDAR data and applied multiple linear regression analysis to estimate fine dead fuel load. LiDAR is a remote sensing technology that uses laser light to measure distances to the Earth's surface, providing highly accurate topographic data. By using LiDAR-derived metrics, the researchers aimed to create models that could accurately predict fine dead fuel load, which is a key factor in wildfire behavior. Model performance was evaluated using several statistical measures: the coefficient of determination (R2), root mean squared error (RMSE), and mean absolute error (MAE). These metrics help determine how well the model predictions match the actual data. A higher R2 value indicates a better fit, while lower RMSE and MAE values indicate more accurate predictions. This study builds on previous research that highlighted the importance of accurate fuel inventory data for effective fuel management planning[3]. Customizing fuel models based on specific forest classes and using hierarchical clustering approaches have been shown to enhance the accuracy of fuel management planning. The integration of site-specific fuel inventory data with predictive models provides forest managers with valuable tools for decision-making. Furthermore, the use of machine learning models to estimate wildfire occurrence has been explored, with an emphasis on making predictions interpretable and intelligible[4]. By employing eXplainable artificial intelligence (XAI) frameworks, researchers have been able to detect regions with a high presence of wildfires and outline the drivers of wildfire occurrence. The integration of such frameworks into decision support systems could support forest managers in preventing and mitigating future wildfire disasters. In conclusion, the study by the University of Bari demonstrates the potential of airborne LiDAR data in accurately estimating fine dead fuel load. By developing predictive models using LiDAR-derived metrics, the researchers have provided a valuable tool for wildfire monitoring, management, and prevention. This approach, combined with previous research on fuel management and machine learning models for wildfire occurrence, offers a comprehensive strategy for addressing the growing threat of wildfires in Mediterranean forests.

EnvironmentEcologyPlant Science

References

Main Study

1) Use of airborne LiDAR to predict fine dead fuel load in Mediterranean forest stands of Southern Europe

Published 4th July, 2024

https://doi.org/10.1186/s42408-024-00287-7

Related Studies

2) Influences of wildfire on the forest ecosystem and climate change: A comprehensive study.

https://doi.org/10.1016/j.envres.2023.117537

3) Developing Custom Fire Behavior Fuel Models for Mediterranean Wildland-Urban Interfaces in Southern Italy.

https://doi.org/10.1007/s00267-015-0531-z

4) Explainable artificial intelligence (XAI) detects wildfire occurrence in the Mediterranean countries of Southern Europe.

https://doi.org/10.1038/s41598-022-20347-9


Related Articles

An unhandled error has occurred. Reload 🗙