InSAR technique for earthquake studies

Abstract

An earthquake could be a devastating disaster that leads to losses of human lives . Direct observation of slip distribution at the earthquake source is almost impossible, therefore indirect measuring methods are commonly applied. The research in this thesis focuses on mapping earthquake affected areas with the Interferometric Synthetic Aperture Radar (InSAR) technique, and utilises the surface deformation observations as input for the slip inversion process in order to model the earthquake. Using InSAR to map earthquakes can obtain dense ground deformation observations with metre-scale resolutions at centimetre accuracy. The deformation of the typical area during a certain time period is observed by repeat-pass InSAR, where in this research is the two-pass Differential InSAR (DInSAR). The displacement map generated from the InSAR result will be the main source for this procedure; the subsequent steps include the forward modelling and slip distribution inversion using the sampled surface deformation observations from the displacement map. These dense observations support the slip inversion using an elastic model named the Okada model. The Levenberg-Marquardt algorithm is applied for nonlinear inversion while the damping value is utilised for linear inversion. Four case studies are presented in this work. The first two are the Sichuan earthquake using ALOS PALSAR data and the L’Aquila earthquake using ENVISAT ASAR data respectively, showing the different capabilities of L-band and C-band mapping of the earthquakes. The third case study is the 2010 Darfield earthquake using both C-band and L-band observations to map and model the event. The fourth case study is the 2015 Gorkha earthquake in Nepal using ALOS-2 data to invert the slip distribution of the earthquake, and a combining inversion with ScanSAR and Stripmap is conducted. This work reveals the ability of L-band and C-band radar interferometry to map a large earthquake area and demonstrates the possibility of using contour expansion for emergency response.