The scope of this dissertation is to provide a discussion about the exploitation of multiple SAR images for the remote sensing of natural scenarios, and describe new procedures to infer information from the data. Due to the variety of scenarios that are commonly imaged by SAR sensors, different kinds of analyses will be taken into account. In the case of forested areas the object under analysis is commonly represented by the vertical structure of the trees, to be inferred through tomographic techniques. The opposite case is given by scenarios for which surface scattering can be assumed. Such scenes are typically investigated through interferometric techniques, in order to estimate the topographic profile and the displacement field undergone in the observation time. A unified treatment of SAR Tomography (T-SAR) and Interferometry (InSAR) will be provided by considering the multi-image SAR system as a a tool for the 3D reconstruction of the upper portion of the Earth''s surface. By virtue of this interpretation, T-SAR and InSAR will be introduced as the result of particular hypotheses about the imaged scene.