The Spatiotemporal Evolution of Granular Microslip Precursors to Laboratory Earthquakes

Abstract

Laboratory earthquake experiments provide important observational constraints for our understanding of earthquake physics. Here we leverage continuous waveform data from a network of piezoceramic sensors to study the spatial and temporal evolution of microslip activity during a shear experiment with synthetic fault gouge. We combine machine learning techniques with ray theoretical seismology to detect, associate, and locate tens of thousands of microslip events within the gouge layer. Microslip activity is concentrated near the center of the system but is highly variable in space and time. While microslip activity rate increases as failure approaches, the spatiotemporal evolution can differ substantially between stick‐slip cycles. These results illustrate that even within a single, well‐constrained laboratory experiment, the dynamics of earthquake nucleation can be highly complex.

Deep Learning Can Predict Laboratory Quakes From Active Source Seismic Data

“…The state variable θ is commonly interpreted as a characteristic contact time, which can be indirectly probed using seismic/ultrasonic data. That is, a longer contact time would increase the interfacial stiffness k I , which in turn would influence both wave velocity v and amplitude A through the following relationships (Tattersall, 1973):…”

Review of studies on the stick-slip behavior of granular fault gouge

“…因此, 体系出现大量的微滑事件可以作为颗 粒断层泥滑动失稳的"先兆". 随着体系滑动失稳的 临近, 微滑活动的频率增加, 但不同粘滑循环之间的 微滑事件时空演化存在显著差异, 反映了真实地震 断层成核机理的复杂性 [35] . 上述物理试验大都采用声发射技术探测体系内 出现的微滑事件, 并用声发射信号的强度表征微滑 事件的大小, 但现有的声发射技术难以将微滑事件 定位到颗粒级别.…”

“…在 这种背景下, 机器学习是一个不错的选择 [97,98] . 由于 其具有自动化、高效率的大数据处理能力, 机器学习 被广泛应用于地球物理领域的诸多方面, 包括地震 震级评估 [99,100] 、地表变形速率预测 [101] 、地震探测和 地震早期预警 [35,102] .…”

Predicting Fault Slip via Transfer Learning

“…Since the plates and particles work together as an ensemble, it is the aggregate energy evolution that governs the stick-slip behavior in granular fault gouge. In the bi-axial experiment, the source of the AE signal is at the grain contact level 42 .…”

“…Fault gouge contacts broadcast AE independently and/or simultaneously 42 , and displace the sideblocks equivalently along the dimensions of the block due to the extreme stiffness of the steel, in analogy to the E k behavior in the simulation. Thus, the E k is approximately equivalent to the magnitude of the continuous AE time series (norm of the acoustic emission recorded by the two channels of the lab experiments), which is the source of elastic waves.…”