Source Parameters Estimation Based On Empirical Green’s Function Spectral Ratio Method

Seismic stress drop is an important parameter to characterize the source characteristics,and the corner frequency of the observed spectrum is the key parameter to calculate stress drop.In recent years,foreign scholars have studied detailedly about source parameters estimation based on telemetered seismic records by empirical Green’s function(EGF)spectral ratio method.EGF spectral ratio method has become increasingly mature in eliminating propagation path attenuation and local site effect of ground motion.However,based on the strong motion records,the generalized inversion method is widely used in China to estimate the source parameters,while the application of EGF spectral ratio method needs to be developed.This paper focuses on exploring the feasibility of applying EGF spectral ratio to strong motion records and improving the stability of source parameter estimating.This paper aims to lay a foundation for the application of EGF spectral ratio method to strong motion record in China.The main work of this paper is as follows:(1)Based on strong motion records,we take Lushan earthquake sequence as an example to verify the feasibility of EGF spectral ratio method to estimate source parameters.The source spectra of medium and small earthquakes(～M4,～M5)are obtained.The selection criterias of EGFs are determined,the amplitude spectrums are obtained by fast Fourier transform,the amplitude spectrum ratio is obtained by Boatwright source spectrum model fitting,and the quality standards of source spectrum are given quantitatively.Finally,corner frequency and stress drop are estimated.The corner frequencies and stress drops of 17 aftershocks are obtained.The corner frequency is mainly in the range of 1.0-3.0Hz,and the stress drop is in the range of 8.2-75.31 MPa,with an average value of 29.00 MPa.The stress drop has no obvious anomaly in time and space.(2)In order to improve the stability of the estimated results,several optimization measures are carried out for EGF spectral ratio method.We take Yangbi earthquake sequence as an example to verify the optimal results,including selecting Yangbi earthquake sequence with higher intensity network and more sufficient small and micro seismic data;selecting the P-wave single phase time series;adding the data processing work such as selecting waveform,picking onset time,intercepting P-wave and SNR requirement;using Multitaper method to obtain amplitude spectrum ratio.We improve the selection criteria of EGFs and the quality standard of source spectrum ratio.The source spectra of small and medium earthquakes(～M3,～M4,～M5)are also obtained by EGF spectral ratio method and Boatwright source spectrum model.Finally,corner frequency and seismic stress drop are estimated.The corner frequencies and stress drops of 28 foreshocks and aftershocks are obtained.The corner frequency is in the range of 2.8-12.0Hz,and the stress drop is in the range of3.85 to 88.35 MPa,with an average value of 32.96 MPa.In terms of time,the stress drops before the main shock are larger than which after the main shock,and the average stress drop showed a decreasing trend over the time.In terms of space,there is no obvious abnormality in stress drop.On the whole,the focal parameters of the two earthquake sequences have the following characteristics: the corner frequency of earthquakes with different magnitudes varies widely.With the larger magnitude,the corner frequency is smaller.The corner frequency of small earthquakes is more discrete.The stress drop varies in the range of several MPa and tens of MPa.There is no obvious correlation between the stress drop and the earthquake magnitude.The stress drop of the same magnitude can vary in the range of two orders of magnitude.(3)In view of the instability of source parameter estimation of Yangbi earthquake sequence,we summary and analyze five types of low-quality observation records.In terms of seismic rupture type,the low-quality observation records mainly affecting results are explained.We obtain the source time function for 23 earthquakes and distinguish rupture types,of which 12 earthquakes are self-stopped rupture earthquakes and 10 earthquakes are forced stopped rupture earthquakes.The analysis shows that the source parameter estimation results of forced stopped rupture earthquakes have lower stability.