A novel methodology for detecting anomalous deformation behaviour from satellite-Synthetic Aperture Radar Interferometry (InSAR) is proposed. The representation of InSAR metadata as embeddings within a deep learning framework (EE-DL) is investigated for modelling the spatio-temporal deformation response. To evaluate the performance of the EE-DL approach in SAR interferometry, we conduct experiments over a mining test site (Cadia, Australia) which has been subject to a tailings storage facility failure. This study demonstrates that EE-DL can detect and predict the fine spatial movement patterns that eventually resulted in the failure. We also compare the results with deformation predictions from a common baseline model, Random Forest.

Ground deformation measurements from Synthetic Aperture Radar Interferometry (InSAR) and ground-based prism mon-itoring are compared to Finite Element (FE) simulation over a collapsed tailings dam. InSAR monitoring captured the complex spatial and temporal variability of TSF deformation that is not be possible from the sparse measurements offered by traditional monitoring. Potentially anomalous deformation over the failed area have been detected from InSAR one year before the failure. The InSAR monitoring data and FE modelling results were in broad agreement with ground based prism measurements, both in terms of magnitude and trends. The results, however, deviate significantly immediately pre-ceding the failure, associated to the construction of buttresses. Finally, this study illustrates the complementarity of remote sensing and geotechnical approaches for the monitoring of tailings dams.