Adjoint tomography of the crustal and upper-mantle seismic structure beneath Continental China

东亚因其独特的陆-陆碰撞、不同类型的板块俯冲、不同类型的含油气盆地形成、广泛的陆内裂谷、板内火山作用和岩浆作用等复杂的构造活动而引起了地球科学家的极大关注。追踪这些构造活动背后的驱动力对于理解古老的大陆根、板块边缘的大陆和海洋板块以及岩石圈下面的地幔流动在动态地球框架中如何相互作用至关重要。S必须获得地球的高清晰度多参数地震图像，以检验不同的构造驱动力假说，如地壳缩短或/和印度板块俯冲导致世界S屋顶、青藏高原、下地幔热柱或/和地幔过渡带(410？650公里深)停滞板块的脱水导致热地幔上涌导致中国和朝鲜交界处的板内火山长白山，以及太平洋板块俯冲和退缩导致东下方岩石圈的伸展、应变和减薄以及日本和南中国边缘海的中国开放。在这项研究中，地震成像将绘制地震记录的完整波形，以呈现更逼真的模型，代表地球的内部属性？S。这一成果不仅将为检验构造形成机制的端元提供独立而有力的地震学约束，而且将揭示岩石圈和上地幔的组成、热状态、流动和组构结构，这是对东亚地质、地球化学、岩石学和大地测量观测的补充。由中国及周边地区非常密集的台阵记录的史无前例的地震波形数据集将用于成像东亚之下的地壳和上地幔结构。该数据集由CER阵列、NECESS阵列、INDEPTH-IV阵列、F-NET和其他全球和区域地震台网的地震记录组成。与传统的基于射线理论的地震成像不同，本文的伴随层析成像充分考虑了三维波的传播效应和偏离射线路径的敏感性。在这个实现中，它利用谱元素方法来精确地模拟波的传播。层析成像方法从3D初始模型开始，该初始模型结合了平滑的径向各向异性地幔模型S362ANI和3D地壳模型Crust2.0(可能最近发布的Crust1.0可以用于更精确和更高分辨率的模型精化)。基于共轭梯度法，利用为每个更新的3D模型计算的3D有限频率核，迭代地最小化旅行时间和幅度失配。大规模并行模拟是在XSEDE超级计算机上进行的，该团队可以通过现有的XSEDE研究分配获得这些计算机。最终的模型将允许仔细检查东亚不同构造单元下的地震结构，并更好地了解地下过程如何塑造地表地质和构造特征。初步研究结果表明，地壳和上地幔的不均质性与喜马拉雅地块、青藏高原、塔里木盆地、鄂尔多斯地块和四川盆地等地表构造单元有很强的相关性。窄板特征出现在过渡区上方的平滑初始模型中。初步模型中的3D波速变化与先前研究中的高频P波和S波层析图像相当或更清晰。将进行额外的迭代，以改进3D径向各向异性模型，并解决3D方位各向异性和衰减问题，这些问题与地幔流变学以及地壳和岩石圈的应变率和流动密切相关。这项提案将支持一名研究生和一名研究科学家。这项研究的结果还将对理解3D波速非均质性与地震孕育过程之间的联系产生更广泛的影响。更现实的高清晰度模型将通过提供更准确的地面动图、更稳健的滑动预测和对未来灾难性地震的有限断层表面的破裂估计来帮助评估地震风险。最终的模型将被存档并分发给地学其他领域的定量研究，如地质学、地球化学、岩石学、地球动力学和火山学。利用赖斯现有的3D可视化实验室设施，地球模型可以很容易地向公众展示，以促进地球科学。最后，这一成果将加强美国和中国之间的科学交流，并拓宽未来潜在的合作。

East Asia has drawn much attention of geoscientists due to its complicated zoo of tectonic activities: unique continent-continent collision of India and Asia, different styles of plate subduction, various types of hydrocarbon-bearing basin formation, widespread intracontinental rifting, intraplate volcanism and magamatism. Tracing the driving forces behind these tectonic activities is vital to understand how the ancient continental roots, continental and oceanic plates at the plate margins, and mantle flows underlying lithosphere interact amongst each other in a dynamic earth framework. It?s essential to obtain high-definition multi-parameter seismic images of the earth?s interior to test different hypotheses of the tectonic driving forces, such as crustal shortening or/and Indian plate underthrusting causing the uplift of ?the world?s roof?, the Tibetan Plateau, lower-mantle plume or/and dehydration from the stagnant slab in the mantle transition zone (410 ? 650 km depths) inducing hot mantle upwelling contributing to the intraplate volcano Changbaishan along the border of China and North Korea, and the Pacific slab subduction and rollback producing extension, transtension, and thinning of the lithosphere beneath the East China and opening of the marginal seas of Japan and South China. Seismic imaging in this study will map full waveforms of seismic records to render more realistic models representing the earth?s interior properties. The outcome will not only provide the independent and robust seismological constrains to test the end members of tectonic formation mechanism, but also shed new light on the composition, thermal state, flow and fabric structure of the lithosphere and upper mantle, which are complementary to geological, geochemical, petrological, and geodetic observations of East Asia.An unprecedented seismic waveform dataset recorded by very dense array stations in China and its surrounding regions will be used to image the crust and upper-mantle structure beneath the East Asia. This dataset is comprised of seismic records from the CEArray, the NECESSArray, the INDEPTH-IV Array, F-net and other global and regional seismic networks. Contrary to traditional ray-theory based seismic imaging, adjoint tomography of this study takes into account full 3D wave propagation effects and off-ray-path sensitivity. In this implementation, it utilizes a spectral-element method for precise wave propagation simulations. The tomographic method starts with a 3D initial model that combines the smooth radially anisotropic mantle model S362ANI with 3D crustal model Crust2.0 (potentially recently released Crust1.0 can be used for more accurate and higher resolution model refinement). Traveltime and amplitude misfits are minimized iteratively based on a conjugate gradient method, harnessing 3D finite-frequency kernels computed for each updated 3D model. The massively parallel simulations are carried out on XSEDE supercomputers available to the team via an existing XSEDE research allocation. The final models will allow close examination of seismic structures beneath different tectonic units of East Asia, and better understanding of how sub-surface processes shaping surface geology and tectonic features. The preliminary results already show strong correlations between heterogeneities in the crust and upper mantle with surface tectonic units, such as the Himalaya Block, the Tibetan Plateau, the Tarim Basin, the Ordos Block, and the Sichuan Basin. Narrow slab features emerge from the smooth initial model above the transition zone. 3D wavespeed variations in the preliminary model are either comparable to or sharper than high-frequency P- and S-wave tomographic images from previous studies. Additional iterations will be carried out to refine the 3D radially anisotropic models, and to resolve 3D azimuthal anisotropy and attenuation, which are closely related to mantle rheology and crustal and lithospheric strain rates and flow. This proposal will support one graduate student and one research scientist. The outcome of this research will also have a broader impact on understanding the linkage between 3D wavespeed heterogeneities and seismogenic processes. The more realistic high-definition models will help assess seismic hazard by providing more accurate ground motion maps, more robust slip predictions and rupture estimations on finite fault surfaces for future catastrophic earthquakes. The final models will be archived and distributed for quantitative studies in other fields of geoscience, such as geology, geochemistry, petrology, geodynamics and volcanology. With the existing 3D visualization lab facility at Rice, the earth models can be readily displayed to the public to promote earth science. Finally the results will enhance scientific communication between US and China and broaden future potential collaborations.