论文部分内容阅读
提出了一种校准地震台网以获取高准确度地震矩张量的方法。该方法基于对大型数据集的联合反演以获取矩张量和台站放大,包括传感器的放大率、数据采集系统的增益因子以及在格林函数模拟中忽略的局部场地效应。该方法能够检测传感器的反向极性、传感器的不正确取向,或由于各台站局部地质条件造成的异常场地效应。使用有不同噪声水平、台站布局和多种震源机制的合成数据对该方法的稳健性和准确度进行了测试。数值模拟证实反演程序代码运行良好并产出稳健结果。测试表明,由适当校准过的地震台网的数据计算出的矩张量显著地更加准确。最后,将该方法应用于捷克共和国西波希米亚的观测数据,以校准该地区运行的由22个地方地震台构成的台网,并准确计算了选出的200个微震的矩张量的双力偶(DC)和非双力偶(非DC)分量。结果表明,该方法是有效的,且能容易地应用于校准其他台网。例如,它能用于反演实验室数据,在这种情况下传感器和岩石样本之间的耦合效应难于定量化;或用于矿山和井下数据,在这种情况下往往不知道传感器的校准和取向。此外,该方法还能应用于涉及获取和解释高准确度矩张量及其双力偶和非双力偶分量的各项研究。
A method of calibrating the seismic network to obtain the seismic moment tensor with high accuracy is proposed. The method is based on joint inversion of large data sets to obtain moment tensors and station amplifications, including sensor magnification, gain factors for data acquisition systems, and local field effects ignored in Green’s function modeling. This method can detect the reverse polarity of the sensor, the incorrect orientation of the sensor or the abnormal site effect due to the local geological conditions of each station. The robustness and accuracy of this method were tested using synthetic data with different noise levels, station layouts and multiple focal mechanisms. Numerical simulations confirm that the inversion code runs well and produces robust results. Tests have shown that moment tensors calculated from properly calibrated seismic network data are significantly more accurate. Finally, the method was applied to the observed data in western Bohemia, Czech Republic to calibrate the network of 22 local seismostats operating in the area and to accurately calculate the moment tensor of the selected 200 microseisms Dual-couple (DC) and non-dual (non-DC) components. The results show that this method is effective and can be easily applied to calibrate other stations. For example, it can be used to invert laboratory data where the coupling effect between the sensor and the rock sample is difficult to quantify, or for mine and downhole data, where sensor calibration and orientation. In addition, the method can be applied to various studies involving the acquisition and interpretation of high-accuracy moment tensor and its dual couple and non-dual couple components.