Forest Structure Estimation by means of TomoSAR in front of weather and seasonal variability

Abstract

When observing a scene, a radar system is sensitive to the morphology of the scatterers, but also to their dielectric properties. Therefore, in the framework of forest monitoring, the presence or absence of leaves, the daily and seasonal water cycle in the different structural elements of the tree, as well as rain may have an influence on the 3D reflectivity retrieved by a TomoSAR system at L-band. However, a consistent forest structure estimation can not vary with these non structural effects. Therefore, a measure of forest structure from TomoSAR data at L-band has been suggested, which discriminates in the 3D reflectivity profiles the parameters only affected by structure related properties. The stability of this estimation is checked here on a temporal series of an airborne campaign, carried out by DLR over a test site in the South East of Germany.