Probabilistic Empirical Green’s Function Method And Ground Motion Simulation Of Lushan Ms7.0 Earthquake

Near-fault strong ground motion is an important data to explore the earthquake source and transmission, as well as an effective way to identify the attenuation characteristics of the ground motion parameters of a certain region for the purpose of aseismic design. Currently, near-fault records that are meaningful for engineering projects are still of a limited amount, meanwhile, engineering projects in need of special anti-quake design, such as nuclear power plants, large-scale dykes and super high-rise buildings that are now being constructed increasingly have clear requirements for acceleration response spectrum and acceleration time series as two parameters to be considered in the engineering design. Therefore, it is of great scientific significance and potential value of application to conduct numerical simulation of near-fault strong ground motion of major earthquakes.One of the methods for synthesizing strong ground motion is the Empirical Green’s function method. It takes the foreshock or aftershock as empirical green functions to synthesize the main earthquake. In previous simulation processes parameters were taken deterministically. However, the uncertainties of those source parameters such as fault geometry (length and width), rise time, the dislocation time function of fault plane, the initial rupture point, share wave velocity and rupture velocity affect the simulation results significantly. In this article, considering the simulated strong ground motion results which are obtained by using empirical Green’s function method are affected by the uncertainties of share wave velocity, small events rise time, rupture propagation velocity and the ratio of fault dimensions between large earthquake and small earthquake, we apply a probability approach to the simulation process in order to determine the four above parameters optimal value and the probability, and then assembly the values using logic tree. Thus we can calculate the strong ground motion results and the probability with different configurations. The application of statistic method in searching optimal value is Probability Empirical Green’s function method. We select the April 20,2013 Lushan Ms7.0 earthquake as the study object. Results show that the PGA, acceleration history and the response spectral of the synthesis fit the recordings well.In addition, we discussed effects of the uncertainties of share wave velocity, small earthquake rise time, rupture velocity and the ratio of fault dimensions between large earthquake and small earthquake to the Lushan earthquake event. By using control variable method we analyzed how much of the parameters affect the results respectively. We found that small earthquake rise time doesn’t affect the results most, whereas the left three parameters make the differences, in which the uncertainty of the ration of fault dimensions between large earthquake and small earthquake have the most obvious impact on the results. Furthermore, the comparisons of biases by different parameter configurations showing that using the Probability empirical Green’s function method may reduce errors to some degree when considering parameter uncertainties.