Speaker
Description
Biomass is the seventh Earth Explorer mission of the European Space Agency. Its primary objective is to reduce uncertainties in global carbon stocks of the terrestrial biosphere as well as the corresponding carbon fluxes caused by land use change (LULUCF, mainly deforestation) and land uptake (mainly forest regrowth). To achieve this goal, Biomass will generate yearly and near-global maps of above-ground dry biomass density (AGBD) from data collected by its polarimetric synthetic aperture radar (SAR) operating at P-band (432-438 MHz). Further innovative components of the Biomass mission are interferometry and tomography which will be used for reducing the disturbing effect of the scattering component originating from the ground level. The focus of Biomass is tropical forests where AGBD data is most urgently needed. Biomass will generate wall-to-wall maps of AGBD across South America, Africa, Asia and Australia on a 200-meter grid.
The baseline algorithm for AGBD from Biomass data is based on linear regression and a power-law model, relating AGBD to P-band radar backscatter at HH, VV and VH linear polarization [Soja et al. 2021]. This semi-empirical model has been developed based on airborne P-band SAR data collected prior to the mission over tropical and boreal forests. Further enhancements of the algorithm are expected based on experiences from the Biomass mission data. The baseline algorithm estimates AGBD by model inversion after training the model parameters on a global training data set using linear regression.
Analysis of Biomass level 1A/B/C and 2A data products have shown significant ground topography effects which cause errors in the AGBD estimates unless mitigated. Results from overlapping ascending and descending tracks clearly show this effect. Airborne SAR data collected prior to the mission launch also shows significant topographic effects [Bennet et al. 2024]. Results from analysis of the Biomass data will be shown at the conference, in particular comparing L2A ground-notched and tomographic data over tropical forested areas. Polarization differences are investigated and interpretations of the effects are provided. Prospects for mitigation will be proposed.
References:
Soja et al., Mapping above-ground biomass in tropical forests with ground-cancelled P-band SAR and limited reference data, Remote Sensing of Environment, vol. 253, 2021
Bennet et al., Sensitivity of P- and L-band SAR Tomography to Above-Ground Biomass in a Hilly Temperate Forest, IEEE Transactions on Geoscience and Remote Sensing, vol. 62, 2024.