Tropical forest biomass recovery using GeoSAR observations

Tropical forests host some 40% of the world\´s above-ground vegetation biomass. Tropical forest biomass estimation from remote sensing is a key issue for REDD and carbon market credit allocation and monitoring. At present there is no consensus on the appropriate remote sensing technologies for tropical forest areas. Cloud cover in the tropics and biomass saturation suggest that a combination of low-frequency SAR and interferometry (either PolInSAR or dual-band interferometric SAR DBInSAR) can provide a solution. The airborne GeoSAR collects X-band and P-band InSAR data simultaneously, at a rate of 288 sq km / minute, and is used for wide-area mapping. Tropical forest biomass recovery using X-P DBInSAR and P-band backscattering cross section has been demonstrated from an airborne platform. The technique is applied to GeoSAR data of tropical forests. We show that GeoSAR X-P interferometric data alone may be used to recover tropical forest biomass, removing ambiguity associated with variation in ground conditions. The effects of terrain slope on biomass recovery are discussed. Airborne observation would yield only a "snapshot" of biomass and carbon stocks. We suggest that a combination of GeoSAR observation with PALSAR data for forest/non-forest classification, plus natural sequestration modelling, should provide an accurate measure of tropical forest biomass temporal variation at high-spatial resolution.