On the difficulties of detecting PP precursors
On the difficulties of detecting PP precursors
复制标题
DOI：
10.1093/gji/ggv105
复制
发表时间：
2015-06
影响因子：
2.8
通讯作者：
S. Lessing;C. Thomas;M. Saki;N. Schmerr;E. Vanacore
中科院分区：
文献类型：
--
作者：
S. Lessing;C. Thomas;M. Saki;N. Schmerr;E. Vanacore
PP前兆是由上地幔过渡带（MTZ）不连续面上的p波下反射形成的地震波。这些地震相位用于绘制不连续地形、锐度和阻抗对比；由此产生的结构变化通常被解释为地幔内部温度和/或矿物学变化的证据。PP前兆和其他地震相已被用于确定410和660公里深度的地震不连续的全球存在。有趣的是，在80%以上的PP前兆观测中，地震波振幅明显弱于地震参考模型预测的振幅。更令人困惑的是，观察到1 - 5%的地震（这是20 - 25%的具有清晰PP波形的地震）即使在存在发育良好的PP波形的情况下，也没有显示出任何来自不连续面的PP前兆的证据。在由数十到数百个事件组成的六个不同的数据集中发现未检测到。我们使用合成模型来检查一套可能导致PP前体缺失的因素。PP与前驱体的起飞角度仅相差1.2 ~ 1.5°；因此，与源相关的复杂性将影响PP和前体。上地幔的PP波衰减会增加PP前体的相对振幅。过渡区内的衰减可以降低前驱振幅，但这将是一种区域现象，仅限于特定的源接收器几何形状。我们还发现很少有证据表明地震射线的理论传播路径偏离预期的分散到达。具有强烈影响的因素包括：在存在大干扰相位的情况下，地震阵列技术中使用的叠加程序，在数十公里量级的不连续面上存在地形，以及在过渡带中速度和密度随深度变化的三维横向非均质性。我们还将观测到的前驱体振幅与相平衡计算的地震模型进行了比较，发现由软锰矿成分推导出的地震速度模型比目前可用的一维地球模型更能再现数据。这主要是由于软锰矿模型在观测到PP前体振幅降低的震中距离上产生了更强的反射系数最小值。为了在一小部分地震中完全抑制前兆，除了变速模型外，还需要其他影响，如局部不连续地形和地震信号处理效应。
The PP precursors are seismic waves that form from underside reflections ofPwaves off discontinuities in the upper mantle transition zone (MTZ). These seismic phases are used to map discontinuity topography, sharpness, and impedance contrasts; the resulting structural variations are then often interpreted as evidence for temperature and/or mineralogy variations within the mantle. The PP precursors as well as other seismic phases have been used to establish the global presence of seismic discontinuities at 410 and 660 km depth. Intriguingly, in more than 80 per cent of PP precursor observations the seismic wave amplitudes are significantly weaker than the amplitudes predicted by seismic reference models. Even more perplexing is the observation that 1–5 per cent of all earthquakes (which are 20–25 per cent of earthquakes with clear PP waveforms) do not show any evidence for the PP precursors from the discontinuities even in the presence of well-developed PP waveforms. Non-detections are found in six different data sets consisting of tens to hundreds of events. We use synthetic modelling to examine a suite of factors that could be responsible for the absence of the PP precursors. The take-off angles for PP and the precursors differ by only 1.2–1.5°; thus source-related complexity would affect PP and the precursors. A PP wave attenuated in the upper mantle would increase the relative amplitude of the PP precursors. Attenuation within the transition zone could reduce precursor amplitudes, but this would be a regional phenomenon restricted to particular source receiver geometries. We also find little evidence for deviations from the theoretical travel path of seismic rays expected for scattered arrivals. Factors that have a strong influence include the stacking procedures used in seismic array techniques in the presence of large, interfering phases, the presence of topography on the discontinuities on the order of tens of kilometres, and 3-D lateral heterogeneity in the velocity and density changes with depth across the transition zone. We also compare the observed precursors’ amplitudes with seismic models from calculations of phase equilibria and find that a seismic velocity model derived from a pyrolite composition reproduces the data better than the currently available 1-D earth models. This largely owes to the pyrolite models producing a stronger minimum in the reflection coefficient across the epicentral distances where the reduction in amplitudes of the PP precursors is observed. To suppress the precursors entirely in a small subset of earthquakes, other effects, such as localized discontinuity topography and seismic signal processing effects are required in addition to the changed velocity model.