The Rise and Fall of the Himalayan-Tibetan Plateau: An Integrated 3-D Finite Element Modeling

摘要：目前对印亚碰撞地球动力学的认识大多基于两种端元模型：基于连续介质力学的平面应力粘性薄板模型和基于模拟实验的平面应变塑性压痕模型。这两个模型都为理解西藏构造提供了重要的框架，但这些模型的二维性和其他简化留下了许多不确定的重要问题。例如，塑性压痕模型预测西藏和东南亚的侧向挤压是适应2000公里碰撞后地壳缩短的主要机制，而粘性薄片模型则强调地壳增厚的作用。这些不同的预测受到模型限制的影响：塑性压痕模型不允许垂直应变，因此不允许地壳增厚，而粘性薄片模型不包括模拟逃逸构造所需的断层和摩擦(塑性)流变学。自20年前引入这些端元模型以来，已经取得了许多改进，但关于地壳增厚和侧向挤压之间的应变分配以及许多其他问题的持续争论表明，需要一种新的方法。拟议的工作是开发一个全三维(3-D)有限元模型来研究喜马拉雅-青藏高原的起伏。通过数十年的密集研究积累的大量多学科观测数据已准备好在更复杂的地球动力学模型中进行整合和解释。由于计算机技术的进步和有限元方法的成熟，三维有限元模型已经成为可能。PI已经建立了原型三维模型，并成功地应用于青藏高原和安第斯山脉的活动构造研究。该项目的主要模型开发包括在有限应变计算中实现走滑断层，并纳入更真实的流变学。有限应变模型将通过使用粘弹性混合流变学来考虑上地壳内的脆性变形，这在粘性薄板模型中通常被省略。瞬时应力和应变率将在具有粘弹性和塑性粘弹性流变学的无限小应变模型中模拟。有限应变和无限小应变模型将被结合起来，以提供对碰撞历史的更完整的模拟。将进行系统的数值实验，以探索印度支那-亚洲碰撞的机制。这一结果将有助于更好地理解这些问题：喜马拉雅-青藏高原地壳变形的主要驱动力是什么？青藏高原是何时、如何隆升的？是什么控制了地壳增厚和侧向挤压之间的应变分配？是什么导致了西藏的扩张？上地壳形变与下地壳形变和地幔岩石圈形变有何关系？通过与美国和中国科学家的合作，数值模拟将与青藏高原正在进行的地质和地球物理研究紧密结合起来，这些科学家包括殷(加州大学洛杉矶分校)、纳贝莱克(俄勒冈州立大学)和弗劳尔(UIC)。所提出的三维有限元模型将是大陆大范围形变数值模拟的一大进步，并可用于其他板块边界。PI将记录计算机代码，并将其提供给公众。为这项研究汇编的多学科数据集将编入地理信息系统数据库，并向研究界提供。这个项目将涉及本科生和研究生；尖端计算与多学科研究的结合为培养新一代地球科学学生提供了一个很好的机会。

Abstract: Much of the current understanding of the geodynamics of the Indo-Asian collision has been based on two end-member models: the plane-stress viscous thin-sheet model based on continuum mechanics, and the plane-strain plasticine indentation model based on analog experiments. Both models have provided important frameworks for understanding the Tibetan tectonics, but the two-dimensionality and other simplifications of these models have left many important issues in uncertainty. For example, the plasticine indentation model predicts lateral extrusion of Tibet and SE Asia to be the dominant mechanism accommodating the 2000 km post-collisional crustal shortening, whereas the viscous thin-sheet model emphasizes the role of crustal thickening. These different predictions are biased by the model limitations: the plasticine indentation model does not allow vertical strain and hence crustal thickening, whereas the viscous thin-sheet model dose not includes the faults and the frictional (plastic) rheology necessary to simulate the escaping tectonics. Since these end-member models were introduced ~20 years ago, numerous improvements have been made, but the continued debate on the strain partitioning between crustal thickening and lateral extrusion and many other questions manifests the need for a new approach. The proposed work is to develop a fully three-dimensional (3-D) finite element model to investigate the rise and fall of the Himalayan-Tibetan plateau. The large volume of multidisciplinary observational data accumulated from decades of intensive studies is ready to be integrated and interpreted in more sophisticated geodynamic models. The 3-D finite element models are now feasible thanks to the advancement of computer technology and maturation of the finite element method. The PI has already built the prototype 3-D model and successful applied it to studies of active tectonics in the Tibetan plateau and the Andes. Major model developments in this project include implementing strike-slip faults in finite-strain calculations and incorporating more realistic rheology. The finite-strain model will include brittledeformation within the upper crust, often omitted in viscous thin-sheet models, by using a mixed viscous-elastic rheology. Transient stress and strain rates will be simulated in infinitesimal-strain models with viscoealstic and plasto-viscoelastic rheology. The finite-strain and infinitesimal-strain models will be combined to provide a more complete simulations of the collision history. Systematic numerical experiments will be conducted to explore the mechanics of the Indo-Asian collision. The results will lead to a better understanding of these questions: What are the major driving forces for crustal deformation in the Himalayan-Tibetan plateau? When and how did the Tibetan plateau uplift? What controlled the strain partitioning between crustal thickening and lateral extrusion? What caused the Tibetan extension? How is the upper crustal deformation related to that in the lower crust and the mantle lithosphere? The numerical modeling will be closely integrated with ongoing geological and geophysical studies in the Tibetan plateau through collaboration with US and Chinese scientists, including Yin (UCLA), Nabelek (Oregon State), and Flower (UIC). The proposed 3-D finite element models will be a major step forward in numerical modeling of large-scale continental deformation and can be used for other plate boundaries. The PI will document the computer codes and make them available to public. The multidisciplinary data sets compiled for this study will be organized into a GIS database and made available to the research community. This project will involve undergraduate and graduate students; the integration of cutting-edge computation with multidisciplinary studies provides a great opportunity for training a new generation of geoscience students.