Integrating satellite imagery with simulation modeling to improve burn severity mapping

Area of Science:

• Remote Sensing
• Ecological Modeling
• Geospatial Analysis

Background:

• Burn severity mapping is crucial for fire science and resource management.
• Existing methods rely on satellite imagery or spatial fire effects models.
• There is a need for more accurate and integrated burn severity mapping techniques.

Purpose of the Study:

• To develop and test a novel integrated mapping approach for burn severity.
• To combine satellite-derived metrics with fire simulation model outputs.
• To improve the accuracy of burn severity maps.

Main Methods:

• Developed a statistical model integrating the Relative differenced Normalized Burn Ratio (RdNBR) with Fire Hazard and Risk Model (FHRM) outputs.
• Validated the integrated model using 285 Composite Burn Index (CBI) plots in Washington and Montana.
• Compared the integrated model's performance against image-based and simulation-based models alone.

Main Results:

• The integrated model explained more variability in CBI (R² = 0.47) than image (R² = 0.42) or simulation models (R² = 0.07) alone.
• The integrated model achieved lower mean squared error (MSE = 0.28) compared to image (MSE = 0.30) and simulation models (MSE = 0.49).
• Maps generated by the integrated model showed the highest overall accuracy (63%).

Conclusions:

• Combining satellite imagery analysis with fire effects simulation models offers significant benefits for burn severity mapping.
• The integrated approach provides more accurate burn severity maps than standalone methods.
• Future improvements in simulation model performance may be achieved with higher quality input data.