Continental and oceanic upper mantle structure from seismic array data

地球上地幔的速度剖面以地震速度的不连续跳变为特征。主要的速度不连续面（或简称不连续面）位于大约410和660公里的深度。这两种不连续性都可以用地幔物质中主要橄榄石成分的固-固相变来解释。尽管如此，地幔物质的次要成分将在不同深度引入额外的、大多是较小的、不连续的速度跳跃。这些转变使上地幔结构的地震图像复杂化。为了探测这些不连续面和对上地幔精细结构进行成像，需要进行高分辨率的研究，这对地幔的矿物物理构成以及地幔动力学和演化的地球动力学建模具有重要意义。我们建议使用位于印度和澳大利亚的地震阵列记录的走时和波形信息来解析澳大利亚北部和印度北部和东部的上地幔结构。主要地震带位于离这些阵列较远的范围内，这使得研究由于不连续面处速度增加而引起的地震波三倍。这些阵列将收集数千次地震记录，以实现对研究区域的密集覆盖。利用时间序列叠加技术，我们能够解决不同分支的三乘法和旅行时间测量精度高。在正演模拟方案中使用这些信息将使我们能够建立上地幔速度结构和不连续面深度位置的模型。此外，叠加技术增加了相干到达的信噪比，使我们能够使用源自上地幔不连续的细微到达的波形信息。我们将使用三次到达和不连续处S-to-P转换的波形建模来解决速度增加的精细尺度结构。一维和高性能的波传播技术将用于模拟不连续面的精细结构对波场的影响。这项研究将对地球不同构造区域（包括大陆-大陆碰撞带和近代海洋俯冲带）上地幔的组成和动力学提供重要的约束。

Velocity profiles of the Earth's upper mantle are characterized by discontinuous jumps of the seismic velocities. The main velocity discontinuities (or simply discontinuities) are located at depths of approximately 410 and 660 km. Both of these discontinuities can be explained by solid-solid phase transitions in the major olivine component of the mantle material. Nonetheless, the minor constituents of the mantle material will introduce additional, mostly smaller, discontinuous jumps of the velocities at different depths. These transitions complicate the seismic image of the upper mantle structure. High resolution studies are necessary to detect these discontinuities and to image the fine scale structure of the upper mantle with strong implications for the mineral-physical constitution of the Earth's mantle and geodynamical modelling of dynamics and evolution of Earth's mantle. We propose to use traveltime and waveform information from data recorded at seismic arrays located in India and Australia to resolve the structure of the upper mantle beneath northern Australia and northern and eastern India. Major earthquake belts are located in a distant range from these arrays that allows the study of the seismic wave triplications due to the velocity increases at the discontinuities. Several thousand earthquakes recorded at the arrays will be collected to achieve a dense coverage of the study area. Using time series stacking techniques we are able to resolve the different branches of the triplication and measure traveltimes with high precision. Using this information in forward modelling schemes will allow us to develop models of the upper mantle velocity structure and the depth location of the discontinuities. Furthermore, stacking techniques lead to increased signal-to-noise ration of coherent arrivals allowing us to use waveform information from subtle arrivals originating from the upper mantle discontinuities. We will use waveform modelling of the triplicated arrivals and of S-to-P conversions at the discontinuities to resolve the fine scale structure of the velocity increases. One-dimensional and high-performance wave propagation techniques will be used to model the effect of the fine-scale structure of the discontinuities onto the wavefield. This study will put important constraints on the composition and dynamics of the upper mantle in different tectonic regions of the Earth including a continent-continent collision zone and recent oceanic subduction.