Physical model of deep low-frequency earthquakes occurred beneath Japan Island Arc

自上世纪70年代S地震以来，日本岛弧东北部火山前锋下方发生了几次深度为20~40公里的地震。到目前为止，已识别的深部低频地震(DLF)的震级几乎都在2.5级以下，因此我们对深部低频地震的信息很少，除了它们的波形和震源位置。我们发展了一种新的算法来估计由同源动力学激发的多个局部时间序列的平均K2O-朗之万矩阵函数，并提出了“平均耗散谱”的新概念。将该算法应用于秋田县2001年7月11日的地震频域分析，成功地将震源振动系统的特征和震源激励过程的特征分别分离为平均耗散项和波动项。连续地震动称为深部低频震颤(f…最近在日本西南部的前弧，沿着俯冲的菲律宾海板块发现了更多。将平均耗散谱方法应用于LFT数据，得到了LFT波形的特征频率和衰减率。我们确认在LFT的主频范围1-5赫兹有几个峰。2003年LFT的衰减率约为0.20,2002年的LFT衰减率约为0.15~0.18。DLF尾波的大部分满足平稳特性，线性和奇次非线性变换的因果值相对较高。这一数量属性在所有DLF中都是常见的。Julian(1994)提出的解释火山地震许多特征的非线性过程可以产生DLF和深部低频地震所代表的因果值的两种特征。当然，这个模型并不是一个孤立的物理过程，导致了两种因果关系的价值特征。然而，这一物理过程对于DLF常见的震源模型和深部低频震颤是有效的候选。较少

Since the 1970's, a few earthquakes with depths ranging 20 km to 40 km have been observed just beneath the volcanic front in the northeastern part of Japan Island Arc. Almost all of the deep low-frequency earthquakes (DLF) recognized up to now have had magnitudes less than 2.5, so that we have little information concerning deep low-frequency earthquakes except for their waveforms and the locations of their hypocenters. We develop a new algorithm to estimate an averaged KM_2O-Langevin matrix function for multiple local time series excited by the same source dynamics, and propose the new concept of "Average Dissipation Spectrum". This algorithm was applied in the analysis of DLF occurred in Akita prefecture on 11 July 2001, and we succeeded in separating the characteristics of the source vibration system and the source excitation process into the averaged dissipation term and the fluctuation term, respectively.The continuous seismic tremor called the deep low-frequency tremor (LFT) was f … More ound very recently in the Southwest Japan fore-arc, along the the subducting Philippine Sea Plate. We applied the method of average dissipation spectrum to the LFT data and obtained the characteristic frequency and decay rate of the LFT waveforms. We confirm the several peaks in the dominant frequency range 1-5Hz of the LFT. The decay rate of the LFT is about 0.20 for the data of 2003 and 0.15 to 0.18 for the data of 2002.The major parts of the coda waves of DLF satisfy the stationary property, and the causality values for the linear and odd-degree non-linear transformations are relatively higher than those for the even-degree non-linear transformations. This quantitative property is common among all DLF. A nonlinear process to explain many characteristics of volcanic tremors proposed by Julian (1994) can produce the two kinds of characteristics of causality values represented in DLF and in the deep low-frequency tremor. Of course, this model is not a solitary physical process causing the two kinds of causality value characteristics. However, this physical process is a valid candidate for the source model that is common to DLF and the deep low-frequency tremor. Less