Integrating Laboratory-Generated Fault Slip Events with Acoustic Wave Measurements

尽管不断取得进步和突破，但大地震仍然是对社会的持续威胁，这主要是由于预测的困难。 观测地球物理学可以提供地震发生的地点和时间的一般信息，但更精确的估计需要对断层运动作为一种地球物理过程有更全面的了解，这可以从实验室实验中收集。最近的突破表明，现实的，板速率驱动速度厘米/年可以揭示断层滑动不稳定的趋势，不仅可能导致地震，而且在自然样品中的大范围的“慢滑”，现象，不能捕捉到更快，更典型的实验驱动率。 在这里，我建议利用摩擦不稳定性和各种类型的滑动事件发生在板率驱动速度的趋势。 对于确定为非稳定的样品，我们可以在这些样品中诱导滑移事件，并且还通过实验室控制的参数（法向应力和装置刚度）操纵这些滑移事件。 我们可以测量这些滑动事件的特性，并测量任何随后的声发射（AE）与一个计划的宽带声学测量系统安装在实验室直接剪切设备。 这有助于实验室中的AE和现场的地震测量之间的直接比较。 AE数据可以与机器学习例程和微观结构分析相结合，以帮助最大限度地识别数据中的模式。 拟议的项目将解决地震科学中的重要问题，如慢地震的性质和有利于快速地震滑动的条件，以及可能识别可用作地震前兆以改善地震预警的力学和声发射数据中的模式。

Despite continual advances and breakthroughs, large earthquakes remain a constant threat to society, largely due to difficulties in prediction. Observational geophysics can provide general information on the where and broadly when earthquakes occur, but more precise estimations require a fuller understanding of fault movement as a geophysical process, which can be gleaned from laboratory experiments. Recent breakthroughs have revealed that realistic, plate-rate driving velocities of cm/yr can reveal a tendency for fault slip instability that may result in not only earthquakes, but also a wide range of “slow slip” in natural samples, phenomena that cannot be captured at faster, more typical experimental driving rates. Here, I propose to leverage the tendency for frictional instability and various types of slip events to occur at plate-rate driving velocities. For samples determined to be frictionally unstable, we can induce slip events in these samples and also manipulate these slip events via lab-controlled parameters (normal stress and apparatus stiffness). We can measure the characteristics of these slip events, and measure any ensuing acoustic emissions (AEs) with a planned broadband acoustic measurement system installed on laboratory direct-shear devices. This facilitates a direct comparison between AEs in the laboratory and seismic measurements in the field. The AE data may be combined with machine learning routines and microstructural analyses to help maximize the identification of patterns in the data. The proposed project will address important questions in earthquake science such as the nature of slow earthquakes and the conditions conducive to fast, earthquake slip, and the possible identification of patterns in the mechanical and AE data that may be used as earthquake precursors to improve earthquake early warning.