Using a dynamic earthquake simulator to investigate controls on slow-slip events, subduction earthquakes, and their interactions

在过去的二十年里，先进的大地测量和地震仪器揭示了沿俯冲带的各种滑动行为。除了在几秒到几分钟内滑动高达几十米的大地震，这些地震会引起强烈的地面震动和/或沿海地区破坏性的海啸之外，缓慢滑动事件会在几天到几年的时间里悄悄地发生，滑动几到几十厘米，只有敏感的仪器才能探测到。观测显示慢滑事件和俯冲地震之间复杂的相互作用。哪些因素和过程控制着这些不同的滑动行为及其沿俯冲带的相互作用？基于物理的数值模型可以说是解决这个问题的最有用的工具。以前的模型主要是为缓慢滑动事件或沿俯冲带的大地震单独建立的，这限制了它们探索两者相互作用的能力。在这个项目中，研究人员将使用并进一步开发一个动态地震模拟器来量化控制慢滑事件、俯冲地震及其在一个统一框架中的相互作用的物理因素和过程。这项研究将促进我们对这两种现象的观测特征及其相互作用的物理理解。随着慢滑事件的频繁发生，对这些事件及其在大地震发生中的作用的进一步了解，可以直接用于评估全球俯冲带（包括卡斯卡迪亚和阿拉斯加俯冲带）未来的地震和海啸危害。高性能计算系统将集中应用于该项目，促进其在自然灾害研究和减少中的应用。这项研究将为今后吸收现有大地测量和地震观测资料的努力奠定基础，以便为地震和海啸灾害的分析和减灾开发特定区域的基于物理的模型。该项目将培养未来的科学家从事这些工作。本研究是利用基于物理的模型来探索控制慢滑事件、俯冲地震及其相互作用的物理因素和过程。研究人员将使用并进一步开发一种动态地震模拟器，该模拟器可以捕捉地震周期的自发动态破裂传播和其他准静态过程，包括成核、震后和地震间过程。动态地震模拟器基于有限元方法，因此可以处理非均质地质介质中几何复杂的断层，包括具有俯冲海底特征的浅倾斜俯冲界面。慢滑事件和俯冲地震将在多个地震旋回中自发出现，并可以在我们的多旋回动力学模型中准确捕获。因此，控制慢滑事件发生和特征、俯冲地震发生及其相互作用的物理因素和过程可以量化。研究人员计划选择北部的Hikurangi边缘作为研究的重点区域。他们还将在项目的后期探索Hikurangi南部边缘和日本海沟。通过在选定的案例研究中再现观察到的慢滑事件、历史地震及其相互作用的特征，研究人员将量化主导这些不同滑动行为及其复杂相互作用的因素和过程。这项研究的结果可用于评估世界范围内俯冲带的地震和海啸危害。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

Advanced geodetic and seismic instrumentations during the past two decades have revealed various slip behaviors along subduction zones. In addition to large earthquakes with slip of up to tens of meters over seconds to minutes that cause strong ground shaking and/or damaging tsunami along coastal areas, slow-slip events occur quietly with a few to tens of centimeters of slip over days to years that can only be detected by sensitive instruments. Observations show complex interactions between slow-slip events and subduction earthquakes. What factors and processes controls these different slip behaviors and their interactions along subduction zones? Physics-based numerical models are arguably the most useful tool to address this problem. Previous models have been largely built separately for either slow-slip events or large earthquakes along subduction zones, limiting their abilities to explore the interactions of the two. In this project, the researchers will use and further develop a dynamic earthquake simulator to quantify physical factors and processes that control slow-slip events, subduction earthquakes and their interactions in one unified framework. This research will advance our physical understanding of observed features in the two phenomena and their interactions. As slow-slip events occur more frequently, improved understanding of these events and their roles in the occurrence of large earthquakes can be directly used to assess future seismic and tsunami hazards along subduction zones worldwide, including the Cascadia and Alaska subduction zones. High-performance computing systems will be intensively used in this project, advancing their usage in natural hazards research and reduction. This research will set a stage for future efforts to assimilate available geodetic and seismic observations to develop region-specific, physics-based models for seismic and tsunami hazards analysis and mitigation. The project will train future scientists to pursue these efforts.This research is to use physics-based models to explore physical factors and processes that control slow-slip events, subduction earthquakes and their interactions. The investigators will use and further develop a dynamic earthquake simulator that can capture both spontaneously dynamic rupture propagation and other quasi-static processes of earthquakes cycles, including nucleation, postseismic and interseismic processes. The dynamic earthquake simulator is based on a finite element method and thus can handle geometrically complex faults in heterogeneous geologic media, including shallowly dipping subduction interfaces with subducted seafloor features. Slow-slip events and subduction earthquakes will emerge spontaneously over multiple earthquake cycles and can be captured accurately in our multicycle dynamic models. Therefore, physical factors and processes that control slow-slip events generation and characteristics, subduction earthquakes generation and their interactions can be quantified. The investigators plan to select the northern Hikurangi margin as the focus area of this research. They will also explore the southern Hikurangi margin and the Japan trench at the later stage of the project. By reproducing observed features of slow-slip events, historical earthquakes, and their interactions in the selected case studies, the investigators will quantify factors and processes that dominate these different slip behaviors and their complex interactions. Findings from this research can be used to assess seismic and tsunami hazards along subduction zones worldwide.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.