The geometrical characteristics and the space-time distribution of the kinematic source parameters of the 21 February 2011, Mw 6.2, Christchurch earthquake, New Zealand, have been inferred through a joint inversion of geodetic and strong-motion data. The geodetic data consist of both Global Position System (GPS), from campaign and continuous stations, and Synthetic Aperture Radar (SAR) interferograms from two ascending satellite tracks. The strong-motion data have been recorded at 10 stations located in the Canterbury Plane, offering a quite good azimuthal coverage of the event.
Before performing the data inversion, several preliminary analyses on individual data-sets have been carried out, in order to find the optimal lay-out for the inversion. In particular, the consistency between GPS and InSAR data was checked and some GPS measurements, characterized by large errors, were excluded from the data-set. The strong-motion analyses were instead primarily addressed, to identify the reliable frequency range to be used, through the analysis of stability of S-wave polarization.
The kinematic rupture model was obtained using the nonlinear joint inversion scheme proposed by Delouis et al. (2000), which is based on the simulated annealing algorithm. In particular, for any sub-source in which the fault plane is discretized, we explore for the direction, duration and amplitude of the slip vector, and for the rupture offset time. The geometry and orientation of the fault plane to be used in the inversion procedure is preliminarily inferred from the analysis of the geodetic data. In order to account for the complex pattern of the superficial deformation data (especially of the InSAR data), we adopted a source model consisting of two partially overlapping fault segments, whose dimensions are 15x11 and 7x7 km$^2$, and striking at 60$^o$ and 10$^o$, respectively. From the data inversion we found a slip distribution for the largest plane characterized by a high slip area, with a maximum amplitude of 4.2 m, localized at NE of the hypocenter. The second fault plane activated with a delay of about 4s, and a maximum slip of 2 m. Moreover, the total Source Time Function considering the two events has a total duration of about 7 s. The global seismic moment resulting from the joint inversion is about 3.0x10$^{25}$ dyne cm (Mw 6.2), with an average rupture velocity of 2.0 km/s, and most of the energy release occurring on the main fault plane. The source parameters resolution is studied using both the single and joint data-sets, and applying the errors analysis for the retrieved kinematic rupture model.