摘要
Currently,the fault sliding speed functions or source–time functions used in theoretical seismogram calculations have only produced theoretical results;that is,the results of the fault speed–time functions have been obtained only from theoretical studies,and have not been validated using the measured data.This paper signifi cantly improves the method for calculating fault sliding speed using the Doppler eff ect,from the following four perspectives.First,the paper theoretically demonstrates how to confirm that the seismic waves received in some frequency bands by two diff erent receivers are emitted by the same source.Second,the paper proposes a method to determine the similarity of seismic waves received by two diff erent receivers in some frequency bands;it applies the relative change at the two receivers in Fourier amplitude standard deviation in some frequency bands,and determines that similar seismic waves are emitted by the same frequency bands.Third,to eliminate the interference of reflected and refracted waves,this study uses fault sliding time S-wave records for data processing.Finally,the paper replaces the long-time Fourier transform with short-time Fourier transform to enhance the positioning accuracy of fault sliding times.Based on these perspectives,the paper systematically summarizes a general methodology for calculating the fault sliding speed using the Doppler effect.This method is employed to calculate the fault sliding speed of the Wenchuan earthquake;the calculations reveal that there is a general correlation between the fault sliding speeds of the Wenchuan earthquake and the seismic moment changes.The results confi rm that the sliding speed of the Wenchuan earthquake fault possesses the characteristics of abrupt change,whereby a sudden increase in the sliding speed is followed by a rapid decrease.Generally,the sliding speed is not fast,and sometimes,no sliding occurs.There are obvious diff erences from the currently used sliding speed functions,such as the Haskell function,bell-shaped func
目前计算理论地震图所用的震源时间函数即滑动速率是理论研究结果,缺少实测资料的证实。本文对以往提出的用多普勒效应计算断层滑动速率进行了几方面的改进,首先在理论上对如何确定两个接收台站接收到的地震波是由震源同一频段地震波发出的进行了阐述;其次提出用两个台站接收到的地震波在某频率段内傅里叶幅值标准差的相对变化确定两者是否具有相似性,相似的地震波是由同一频段地震波发出的;再次,为消除反射波和折射波的干扰,仅选用S波在断层滑动时间内的地震记录进行数据处理;最后提出用短时傅里叶变换代替长时傅里叶变换提高了断层滑动速率时间定位的准确性。在上述工作的基础上,系统总结了用多普勒效应计算断层滑动速度的一般方法。用上述方法计算了汶川地震断层滑动速率,计算结果的地震矩释放过程与反演结果基本吻合,并证实了汶川断层滑动速率具有突变的特征,即滑动速率突然增大、迅速下降的特点,绝大多数时间滑动速率不大,有的时间段断层没有滑动。这与目前经常使用的滑动速率函数有明显的不同。用多普勒效应计算断层滑动速率,无需掌握目前不太清楚的地壳介质信息,仅仅需要知道地震记录、震中位置和记录台站位置信息,计算方法物理意义清晰、所需参数容易获得。用多普勒效应计算断层滑动速率丰富了断层滑动速率计算理论。
基金
supported by the Educational Department Project of Liaoning province:LJYL040(No.551610001219)。
