The Research Of Source Rupture Process Inversion Method Based On Source Scanning Algorithm

A perfect inversion of rupture process of earthquakes should have the advantages of high resolution, strong stability, good objectivity and computational efficiency should be considered. But the result obtained by current methods can't meet these requirements.Considering the earthquake rupture process as a process of a number of smaller events, the inversion will be equivalent to locating these smaller events and determine their origin time. So, the inversion of earthquake rupture process can be completed using Source-Scanning Algorithm (SSA). The method was proposed by Kao and Shan in 2004 and it uses waveform to locate and image source distribution. SSA uses a so-called "brightness function" to calculate and determine whether there is a hypocenter at a given space-time point. It doesn't require priori assumptions on the source area. The whole process needn't human intervention, so the method can ensure the maximize objectivity and stability of the results.In order to improve the resolution of SSA, theτ-p transform was introduced into the original method to separate the surface wave. By taking advantage of the surface wave's low and constant velocity characterization inτ-p coordinate transform, SSA can achieve the precise characterization of the rupture process.In order to test the sound effect of improved SSA on the inversion of earthquake rupture process, a theoretical model was constructed. The model is composed of an earthquake rupture zone and a seismic network. The rupture was divided into 9 sub-events and the seismic network with 8 Seismic stations offered the surface wave records for us. Those surface wave records can be generated by a surface wave simulation method. After a pulse-based waveform procedure, the records can be scanned by SSA, and the outputs can accurately recover the origin times, the hypocenters and the initial dislocation values of all sub-events. The algorithm shows good stability and objectivity in the inversion results without any human intervention during the calculating.The values of the "brightness function" obtained by improved SSA are the initial dislocation values of the sub-events. And this is equivalent to that the distribution information of the sliding displacement in each section of the rupture zone has been given directly. The results show that SSA could be used in earthquake rupture inversion and is of high resolution, strong stability and good objectivity.