Anisotropic structure of Earth's inner core from noise correlations

地球固体内核是由液态外核在地球冷却时结晶而成的。内核具有轴对称各向异性，其快轴近乎南北排列，这是由于内核铁晶体在这一方向上的优先取向。以前对内核结构的研究已经显示出非常复杂的情况。因此，了解这种突出的各向异性的起源是当今地球科学的基本问题之一；而成像内核复杂的三维各向异性结构在这一努力中至关重要。不幸的是，地球深处的地震成像受到地震和台站位置分布的限制。这个项目试图探索一种新的技术，根据地震噪声(从周围环境和地震尾波)的相关性提取出一组截然不同的地球深处观测数据。我们预计，这种改进的内核地震模型将使矿物学和地球动力学社区受益，因为它对他们的模型提供了约束。该项目对理解地球的各向异性的起源、组成、动力学和演化具有重要意义，特别是对S核和地球内核的各向异性的理解具有重要意义。深穿透波的反演克服了传统地震方法的一些缺点，将对揭示地球深部结构产生根本性的影响--S。该项目建立在最近从环境噪声或地震尾波的相关性中提取体波的成功经验的基础上。在过去的十年里，环境噪声相关技术被广泛地应用于地球结构的研究，特别是岩石圈结构的面波层析成像。最近的研究表明，从环境噪声或地震尾波的相关性中提取体波是可行的。最近，我们成功地从单台的自相关或不同台站记录的数据之间的互相关中，利用经验格林？S函数组同时反演了清晰的三重PKP相和核心到达波PKIKP2和PKIIKP2。这些观测使我们能够利用不同阶段之间的不同旅行时来消除或减少地幔不均一性和震源误差的影响。新的PKIKP2和PKIIKP2数据表明，地球内核存在地震各向异性，其快轴在赤道附近排列，而各向异性的形式与在外核观察到的南北排列的各向异性不同。在这个项目中，我们试图探索这些新类型的数据来约束内核的3D各向异性结构。噪声互相关的不同分支之间的不同pkp旅行时间提供了与现有地震数据截然不同的数据覆盖范围。不同的PKIIKP2-PKIKP2旅行时间提供了内核最深部分的全新样本。具体地说，我们将系统地检索PKIKP2和PKIIKP2，同时对世界各地的所有可用阵列应用自相关和互相关，并通过对距离大于145度的站阵列对应用互相关来系统地检索三个重复的PKP分支。然后，我们将把新的基于噪声的数据与地震数据结合起来，对3D内核各向异性结构进行建模和反演。

The Earth's solid inner core is formed from the crystallization of the liquid outer core as the Earth cools. The inner core possesses axisymmetric anisotropy with its fast axis aligned nearly North-South due to the preferred alignment of the inner core iron crystals in this direction. Previous studies of the inner core structure have shown great complexities. Thus, understanding the origin of this prominent anisotropy is one of the fundamental questions of earth science today; and imaging the complex three-dimensional anisotropic structure of the inner core is crucial in this endeavor. Unfortunately, seismic imaging of the deep Earth has been limited by the distribution of earthquake and station locations. This project seeks to explore a new technique that extracts a vastly different set of observations for the deep Earth based on correlations of seismic noise (from ambient environment and from earthquake coda). We expect that this improved inner core seismic model will benefit the mineralogical and the geodynamical communities, as it provides constraints on their models. The project has implications for understanding the origin of the anisotropy, composition, dynamics, and evolution of Earth?s core and the inner core in particular. The retrieval of the deep penetrating waves overcomes some of disadvantages of traditional earthquake-based methods, which will have a fundamental impact on revealing the structure of the Earth?s deep interior.This project builds on recent successes in extracting body waves from correlations of either ambient noise or earthquake coda. During the last decade, ambient noise correlation technique has been widely used to study the structure of our planet, in particular surface wave tomography of lithosphere structure. More recent studies have shown feasibility to extract body waves from correlations of ambient noise or earthquake coda. We have recently succeeded in retrieving clear triplicated PKP phases and inner-core arrivals PKIKP2 and PKIIKP2 simultaneously using stacks of empirical Green?s functions from autocorrelations of single station or cross-correlations between data recorded at different stations. These observations allow us to use differential travel times between different phases to eliminate or reduce the influence of mantle heterogeneity and source errors. The new PKIKP2 and PKIIKP2 data suggest a seismic anisotropy in the inner-inner core of the Earth that has a fast axis aligned near the equator and a different form of anisotropy from the North-South aligned anisotropy observed in the outer inner core. In this project, we seek to explore these new types of data to constrain the 3D anisotropic structure of the inner core. The differential PKP travel times between different branches from noise cross-correlations provide vastly different data coverage from existing earthquake data. The differential PKIIKP2- PKIKP2 travel times provide brand new samples of the deepest part of the inner core. Specifically, we will systematically retrieve PKIKP2 and PKIIKP2 simultaneously applying autocorrelations and cross-correlations to all available arrays around the world and systematically retrieve triplicated PKP branches by applying cross-correlations to pairs of station arrays at distances greater than 145 degrees. We will then combine the new noise-based data with earthquake data to model and invert for 3D inner-core anisotropic structure.