عنوان مقاله [English]
For large earthquakes, rupture characteristics including rupture velocity and fault extension are important parameters that reflect the fault properties and complexities. One of the most important tasks for earthquake monitoring agencies is to determine a finite source rupture model as quickly as possible so that a map of regions with the strongest shaking can be provided to guide emergency response and rescue. In many cases, the epicenter is not the most severely damaged region. One of the recently used methods to image the source and rupture details is back-projection (reverse time migration), which has some advantages comparing to traditional methods such as finite-fault source inversion; since it is much faster (the computation is relatively easier than inversion) and it can be applied to different frequency bands, even high frequencies, and the only a priori information required is a radial velocity model and a hypocentral estimate. In this method, seismic arrays at teleseismic distances are used. Since the back-projection technique is sensitive to the array geometry, array response function (ARF) is used to choose the array with the least artifact. In order to compute the ARF, the process is the same except the fact that the synthetic seismograms are used instead of real seismograms. To investigate the rupture propagation and energy release of two earthquakes, 2007/06/18 Mw 5.9, Kahak, and 2010/09/27 Mw 5.5, north of Kazeroon, a back-projection of teleseismic P-wave with X4 (China) and YP (northeast China) seismic network arrays, vertical component data high-pass filtered at 1.0 Hz are used. It is assumed that the first part of the seismograms is due to the failure at hypocentre and later parts come from rupture front. To determine the rupture propagation that is necessary to know which point in source area has caused the radiation of energy, a grid of points in source area is set. This grid covers most of the aftershocks region. The back-projection analysis used in this study does not have very good depth resolution, so that grid is 2-Dimensional and the depth of grid is constant; hence, the waveforms are stacked at every time window for all grid points and the back-projection method determines which grid points are the source of seismic radiation in each time window of the teleseismic P waves. In this method, seismograms are stacked for grid point to obtain a direct image of the source. Stacking procedures sums the energy that is radiated from the grid point constructively and cancels out other energy patterns present in the seismograms. Resulting maps show the squared amplitudes of the stacks, which are proportional to the radiated high frequency seismic energy. According to the results, for Kahak earthquake, the rupture is in order of 1.9±0.006 km-1 and the rupture front propagates southwest to northeast about 8±1 seconds. For north of Kazeroon earthquake, the rupture velocity is 1.6±0.003 km and the total time of propagation is 15±1 seconds. The back-projection method is usually used to determine slip distribution of large earthquakes using a very dense array. However in this study we show that the back-projection method can even be extended to study moderate size earthquakes.