Analysis of the reliability and resolution of the earthquake source history inferred from waveforms, taking the Chi-Chi earthquake as an example

Abstract

In the inversion of observed waveforms to obtain the earthquake source faulting process some uncertain factors may influence the reliability of the results. Several numerical tests have been analysed in this paper in order to understand the effect of some of these uncertain parameters, such as the number and width of multiple time windows that define the source-time function, the assumed strike and dip of the fault, the velocity structure, and noise. Also, the influence of the sampling rate, the time window of the data and the distribution of observational stations is studied. Our research indicates: The near-source waveform data can adequately recover the source rupture process of the shallow part of the fault. However, they poorly constrain the slip distribution on the deep part of the fault. Using near-source and teleseismic waveform data jointly can provide a more complete view of the rupture process of the whole fault. The result using near-fault data is very sensitive to the assumed strike and dip parameters of the fault model. A 2° deviation of the presumed strike and a 5° deviation of the presumed dip from that of the true fault will lead to a distorted inversion result. The influence of the uncertainty of the seismic velocity structure for the Chi-Chi earthquake might be ignored if we can take the difference between Ma et al.'s model and Wu et al.'s model as the measure of the media uncertainty.