Application of DInSAR techniques to the monitoring of ground deformations

Co-tutela Universitat Politècnica de Catalunya i Università degli studi di Napoli “Federico II” The aim of the present thesis has been to test the applicability of the innovative Advanced DInSAR techniques in the natural risk mitigation related to subsidence phenomena. In particular, two test sites have been chosen, both located within alluvial plains and affected by subsidence phenomena: Telese Terme (Italy) where no monitoring network has been installed in spite of the great amount of damaged buildings located in the urban area; Murcia city (Spain) where subsidence has caused damage to structures and infrastructures with an estimated cost of more than 50 million euros. In this second case, the institutions have required studies since '90. For this reason, 20 years of monitoring data are available which have allowed the implementation of an integrated monitoring system based upon satellite DInSAR, conventional field techniques and geotechnical data. Therefore these two areas have been chosen to test different approaches in the use of DInSAR results which can complete a monitoring network where available (as in the case of Murcia city) and replace it where it does not exist (as in the case of Telese Terme). In Murcia case study, the correlation between the temporal evolution of ground surface displacement measures (radar and in situ) and the piezometric groundwater level variation has been analysed to determine mechanisms and critical states of failure; this has permitted to implement a finite element model (FEM) of the phenomenon. Therefore, two models have been carried out: one (called "deep") up to the end of the gravel layer (where the pumping takes place) and one (called "shallow") up to the extensometers' base. The results of the deep model have been compared with DInSAR displacements time series which represent the whole deformation of the stratigraphic column. These comparisons have allowed the individuation of local anomalies of the stiffness values, and have permitted a best model calibration. Moreover, the shallow model results have been compared with the extensometers measurements. These comparisons have showed the occurrence of vertical anisotropies of the permeability. This hypothesis has been verified, analysing the available Lefranc's tests and the most detailed stratigraphic columns and a new model has been proposed. The geotechnical model results have been interpolated through the Ordinary Kriging Radar Errors (OKRE) technique. The achiev