3D Monitoring of forest structure by means of Synthetic Aperture Radar Tomography

Synthetic Aperture Radar (SAR) signals at low frequencies can penetrate forest. This capability allows, through advanced imaging techniques such as tomography, the extraction of relevant information related to the three dimensional structural parameters of the observed scene. More specifically, through repeated passes of the sensor, a scene is observed under slightly different observation angles and this angular diversity can be exploited to retrieve the three dimensional reflectivity signature of the scene. Furthermore, by employing different transmitting and receiving polarizations, the polarimetric behavior of the different scatters can be discriminated, allowing an enriched characterization of the scene. Recently, several airborne tomographic SAR experiments at low frequencies have already proven the unprecedented potential of this technique in order to provide a deep insight on forest structure. And, in the next decade, future space missions (BIOMASS, Tandem-L) will allow the monitoring of 3D forest structure, at global scale. However, research is still ongoing to translate this three dimensional reflectivity information into geophysical information that practitioners are familiar with and can assimilate in their analyses. The major drawback comes from the fact that in a tomographic SAR system, the measured signal in a given resolution cell results from the nonlinear combination of the contributions of an undefined number of tree parts (canopy, trunk, leaves …). Therefore, the parameters commonly employed in ecology relying on individual based measures, e.g. tree basal area, diameter of the tree crown, diameter of the tree stem, density of trees among others, cannot be retrieved in a direct way and indirect correspondences between radar measures and forest characterization based on terrestrial observations have to be explored instead. This paper suggests a correspondence between tomographic SAR measures and a set of indices employed to quantify horizontal and vertical forest structure from terrestrial data. More specifically, the correspondence is set through the definition of two complementary descriptors, estimated from the vertical reflectivity profiles, accounting for their simultaneous horizontal and vertical variability. These descriptors are tested on a real scenario (Traunstein forest), in which airborne tomographic SAR acquisitions as well as terrestrial inventory plots and stand information (growth stage and species composition) are available. An extensive cross check between forest structure estimation obtained from remote sensing SAR measures and forest structure estimation from terrestrial observations has been carried out and results show a good correspondence