Precision Influence Analysis Of The Two-stochastic Empirical Green's Function Method In Ground-motion Simulation

The two-step stochastic empirical Green's function method is one of the most effective method to simulate ground motions,the advantages of this method are mainly reflected in the following two aspects:on the one hands,in view of the small earthquake records contain the informations that hard to obtain,such as the seismic source,propagation path and site effect,this method makes full use of rich resources of small earthquakes to simulating the target earthquaks;On the other hands,two-step stochastic empirical Green's function method can ignore the complicated process of the source motion characteristics,and generate the target earthquake by rupture randomly.We can use abundant small earthquake resources to simulating the distructive earthquakes which may happen in particular region,and then research the corresponding shockproof design on the purpose of disaster prevention and reduction.But it also has some shortcomings,such as fault distance,epicentral distance and between the stations and epicenter have the effects on the synthesis of accuracy,even the stations which located on the hanging wall and foot wall can affect the outcome.the Based on these theoretical basis,this paper makes further research through some destructive seismic data resources in recently years and conclusions are as follows:1.Taking the 2016 Kumamoto earthquake as an example,we selected two earthquake which their moment magnitudes are 6.0 and 4.5,then selected 18 trigger stations around them as the research object.The simulation results show that the for most stations the synthetic results match the observed results very well,two-step stochastic empirical Green's function method is feasible,but there are still some limitations in simulating the long period.2.Taking the 2008 Iwate earthquake as an example,we selected the stations which located within 30km of the fault then divided by hanging wall and foot wall as the research object.Based on the goodness-of-fit?Fg?as evaluation criteria.Throughing simulating all of the stations,we can found that the simulation situation of stations which located in foot wall are more precise than that in hanging wall.3.Taking the 2016 Kumamoto earthquake as an example,we selected all of the trigger stations as the research object.Based on the goodness-of-fit?Fg?as evaluation criteria.Throughing simulating all of the stations,we can found that the accuracy of the stations which fault distance and epicentral distance between 20 and 100km is relatively ideal,however,the accuracy can be reduced when the distance is too close or too far,the impact of the fault angle is negligible.4.Taking the New Zealand earthquake,Kumamoto earthquake as examples,By controlling the epicentral distances from the source to trigger station are similar with each other,we selected them which in the same site type as the research object.However,when simulating the Kumamoto earthquake and Jiuzhaigou earthquake,Comaring the observed values with the simulated values,it can be found that it is also feasible to apply the empirical Green's function method outside the rigion.Then based on this conclusion,we used the Kumamoto small earthquakes to simulating the Jiuzhaigou earthquake.we chose the stations with their V_S3030 are similar with each other,due to the different classified evidences of site type.Comparing the observed value with the simulated value,it can be found that the overall synthesis accuracy of V_S3030 in similar degree is higher than that staions in the same site type.