Collaborative Research: Controls on Termination of Great Earthquakes in a Restraining Double-Bend of the Altyn Tagh Fault

由几何复杂性定义的断层分割，如抑制弯曲和台阶，可能在阻止地震破裂方面发挥重要作用，从而限制地震的最大规模。在如此复杂的情况下，控制断层滑动的机械过程仍然难以捉摸。本研究结合野外观测和破裂模拟，对中国西北阿尔金塔格断层200公里长阿克塞抑制双弯曲和台阶的走滑地震不破裂的条件进行了评估。这个弯曲被左侧Altyn Tagh断层系统内的两条断层链所包围，并在两条断层链之间转移滑动。由于阿克塞弯曲与相交的断层隔离，它的行为不太可能受到来自活动断层系统其他部分的应力传递的影响，因此使它成为一个理想的天然实验室，可以将模型结果与现场观察结果进行比较。这项研究将测试基于断层的地震危险性评估，该评估依赖于断层分割，并有助于理解永久变形是如何在断层和周围地壳之间划分的。迄今为止，断层分割的定义主要是通过专家评估来完成的，没有具体的物理基础。数值破裂模拟为断层分割提供了急需的物理基础，但也有很大的局限性。这些限制包括地震前的应力状态是未知的，在几何复杂的区域不太可能是光滑的。因此，数值破裂模型在地震灾害制图中的适用性还有待检验。多地震周期破裂模型将同震破裂与地震间断层外应力松弛相结合，揭示了构造复杂带的应力状态、地震破裂大小和破裂程度模式。该项目将根据古地震学、断层滑动率和断层滑动方向等现场观测来测试这些模式。同时，该研究将推动这些数值模型的发展，使其包括离断层塑性、倾斜和三维断层几何。通过将现场观测和数值破裂模拟相结合，本研究跨越学科界限，努力改变对地震破裂过程及其在地质记录中的表现的认识。在这样做的过程中，该项目建立在国际合作的基础上，共同的基础科学研究目标（了解断层构造过程）和共同的社会需求（了解地震）。这项研究的结果将直接影响对大地震对主要断层的危害的认识。例如，所获得的见解将与了解最南端的圣安德烈亚斯断层破裂造成的潜在地震和震动危险直接相关。

Fault segmentation, defined by geometric complexities such as restraining bends and stepovers, may play an important role in arresting earthquake rupture and thus limiting the maximum size of earthquakes. Understanding the mechanical processes that control fault slip within such complexities remains elusive. This study integrates field observations and rupture modeling to assess the conditions under which strike-slip earthquakes fail to break across the 200 km-long Aksay restraining double bend and stepover of the Altyn Tagh fault in northwestern China. This bend is flanked by, and transfers slip between, two fault strands within the left-lateral Altyn Tagh fault system. Because the Aksay bend is isolated from intersecting faults, its behavior is unlikely to be affected by stress-transfer from other parts of the active fault system, thus making it an ideal natural laboratory to compare model results against field observations. This research will test fault-based seismic hazard assessments that depend upon fault segmentation, and contribute understanding of how permanent deformation is divided among faults and the surrounding crust. To date, defining fault segmentation has been performed largely by expert assessment, with no site-specific physical basis. Numerical rupture simulations offer a much-needed physical foundation for fault segmentation, but with significant limitations. These limitations include that the pre-earthquake stress-state is unknown and unlikely to be smooth in zones of geometric complexity. Thus, the applicability of numerical rupture models to seismic hazard mapping remains untested. Multi-earthquake cycle rupture models that combine coseismic rupture with interseismic off-fault stress relaxation reveal patterns of stress-states, earthquake rupture sizes, and rupture extents in zones of structural complexity. This project will test these patterns against field observations including paleoseismology, fault slip-rates, and fault-slip direction. Simultaneously, this research will advance the state-of-the-art of these numerical models to include off-fault plasticity, and dipping, three-dimensional fault geometries.By integrating field observations and numerical rupture modeling, this research cuts across disciplinary boundaries in an effort to transform understanding of the earthquake rupture process and its manifestation in the geologic record. In so doing, this project builds upon an international collaboration with shared basic-science research objectives (to understand tectonic processes along faults) and shared societal need (to understand earthquakes). The outcomes of this research will directly affect understanding of hazards from great earthquakes on major faults. For example, the insights gained will be of direct relevance to understanding potential earthquakes and shaking hazards from rupture of the southernmost San Andreas fault.