Mapping tropical forest aboveground biomass using airborne SAR tomography

Area of Science:

• Remote Sensing
• Forestry
• Biomass Estimation

Background:

• Accurate mapping of tropical forest aboveground biomass (AGB) is crucial for quantifying land use change emissions and climate mitigation strategies.
• Current remote sensing methods face challenges in precisely measuring AGB in dense tropical forests.

Purpose of the Study:

• To evaluate the capability of P-band tomographic Synthetic Aperture Radar (TomoSAR) for mapping AGB in dense tropical forests.
• To compare different TomoSAR reconstruction algorithms and identify optimal variables for AGB estimation.
• To validate AGB estimation models using independent airborne LiDAR and field inventory data.

Main Methods:

• Compared three TomoSAR reconstruction algorithms: back-projection (BP), Capon, and MUltiple SIgnal Classification (MUSIC).
• Developed univariate and multivariate regression models using TomoSAR variables (e.g., backscattered power, forest height) to estimate AGB at 4-ha grid cells.
• Validated AGB estimates using TropiSAR airborne campaign data, inventory plots, and airborne LiDAR measurements.

Main Results:

• BP-based TomoSAR variables yielded superior AGB estimates compared to Capon and MUSIC.
• TomoSAR-derived forest height and AGB estimation showed an average relative uncertainty below 10% with negligible systematic error.
• Backscattered power at 30m height (HV polarization) was the most accurate single predictor of AGB, outperforming LiDAR metrics; combining it with forest height improved accuracy to <7%.

Conclusions:

• P-band TomoSAR data from the BIOMASS mission offers a novel and accurate capability for mapping tropical forest AGB.
• The combination of backscattered power and forest height from TomoSAR provides a robust method for precise biomass assessment.
• This approach is vital for improving carbon emission quantification and climate change mitigation strategy evaluations.