Conference Agenda

Monitoring the 3D structure and biomass of tropical forests, particularly the Amazon, is essential for global carbon accounting and climate modelling, yet remains a principal challenge for remote sensing. Current spaceborne systems, such as the C-band (5.6 cm) radar on Sentinel-1, are hampered by rapid temporal decorrelation; the short wavelengths scatter from the dynamic canopy surface (leaves and small branches), rendering interferometric analysis ineffective over dense vegetation.

Here we show that the BIOMASS mission's P-band (69 cm) SAR overcomes this fundamental limitation, providing transformative coherence over the Amazon rainforest. We observed a mean interferometric coherence of 0.85 (3-day repeat) for BIOMASS, a 3-fold improvement over Sentinel-1's 0.28 (12-day repeat).

Most strikingly, 93.5% of P-band pixels maintain coherence above the 0.7 threshold required for robust interferometric applications, compared to only 0.8% for C-band. This 117-fold increase in viable data stems from P-band's ability to penetrate the volatile upper canopy and interact coherently with the stable structural framework of the forest—the trunks and ground.

This unprecedented coherence stability unlocks capabilities previously impossible over tropical forests, including the retrieval of forest height via Polarimetric InSAR and the mapping of 3D forest structure through SAR tomography. As the only P-band SAR in space, BIOMASS provides an irreplaceable dataset, filling a critical observational gap for climate science and the implementation of international climate agreements.

The authors acknowledge the support in part from the Center National d'Etudes Spatiales/Terre, Ocean, Surfaces Continentales, Atmosphere (CNES/TOSCA - projects GEDITOMO3D, BayesTomo, BIOMALT, and TS4Biomass). We thank ESA for providing BIOMASS data through the CAL/VAL activities in the project GEDI-TOMO-PP0104473 and AQUABIO_PP0106230.