Transtensional deformation of Montserrat revealed by shear wave splitting
Transtensional deformation of Montserrat revealed by shear wave splitting
复制标题
DOI：
10.1016/j.epsl.2015.06.006
复制
发表时间：
2015-09
影响因子：
5.3
通讯作者：
A. Baird;J. Kendall;R. Sparks;B. Baptie
中科院分区：
文献类型：
--
作者：
A. Baird;J. Kendall;R. Sparks;B. Baptie
在这里，我们调查地震各向异性的上地壳在苏弗里耶尔山火山附近使用剪切波分裂（SWS）分析火山构造（VT）事件。苏弗里耶尔山位于小安的列斯群岛的蒙特塞拉特岛上，1995年开始活跃，此后一直在喷发，有五个主要的喷发活动阶段。我们使用1996年至2007年期间部分跨越三个挤压阶段的地震仪网络记录的数据。地壳中的剪切波分裂通常被认为是由结构特征或应力对齐的裂缝控制的。在这种情况下，快剪切波（横波）的偏振方向将平行于结构的走向，或平行于最大压应力方向。以前的研究分析SWS在该地区使用区域地震观察到的时间变化，这被解释为是由应力扰动与增压的堤坝。我们的分析使用浅得多的源，因此只对地壳上部几公里的各向异性进行采样，没有显示出明显的时间变化。然而，不能排除时间效应，因为在研究期间VT事件发生率的大幅波动以及地震网络配置的变化使其难以评估。平均延迟时间约为0.2秒，类似的幅度报告的更深的板事件，表明大部分的各向异性是在地壳浅。我们观察到明确的空间变化的各向异性，我们认为是一致的结构控制的各向异性造成的一个横向transstensional阵列的故障，穿过火山岩复杂。
Here we investigate seismic anisotropy of the upper crust in the vicinity of Soufrière Hills volcano using shear wave splitting (SWS) analysis from volcano-tectonic (VT) events. Soufrière Hills, which is located on the island of Montserrat in the Lesser Antilles, became active in 1995 and has been erupting ever since with five major phases of extrusive activity. We use data recorded on a network of seismometers between 1996 and 2007 partially spanning three extrusive phases. Shear-wave splitting in the crust is often assumed to be controlled either by structural features, or by stress aligned cracks. In such a case the polarization of the fast shear wave (ϕ) would align parallel to the strike of the structure, or to the maximum compressive stress direction. Previous studies analyzing SWS in the region using regional earthquakes observed temporal variations inϕwhich were interpreted as being caused by stress perturbations associated with pressurization of a dyke. Our analysis, which uses much shallower sources and thus only samples the anisotropy of the upper few kilometres of the crust, shows no clear temporal variation. However, temporal effects cannot be ruled out, as large fluctuations in the rate of VT events over the course of the study period as well as changes in the seismic network configuration make it difficult to assess. Average delay times of approximately 0.2 s, similar in magnitude to those reported for much deeper slab events, suggest that the bulk of the anisotropy is in the shallow crust. We observe clear spatial variations in anisotropy which we believe are consistent with structurally controlled anisotropy resulting from a left-lateral transtensional array of faults which crosses the volcanic complex.