Acoustic emission dataset-1 for experimental study on fluid-driven fault nucleation, rupture processes, and permeability evolution in Oshima granit

Published: 2 May 2024| Version 1 | DOI: 10.17632/grpxc5fty2.1
Xinglin Lei


This is first data set for key experimental data of selected experiments reported in X. Lei (2024), and the second data set is also stored at Mendeley Data (doi: 10.17632/ct25dfrns3.1). Summery of X.Lei (2024): This study investigated the fault nucleation and rupture processes driven by stress and fluid pressure in fine-grained granite by monitoring acoustic emissions (AEs). Through detailed analysis of the spatiotemporal distribution of the AE hypocenter, P-wave velocity, stress-strain, and other experimental observation data under different confining pressures for stress-driven fractures and under different water injection conditions for fluid-driven fractures, it was found that fluid has the following effects: 1) complicating the fault nucleation process, 2) exhibiting episodic AE activity corresponding to fault branching and the formation of multiple faults, 3) extending the spatiotemporal scale of nucleation processes and pre-slip, and 4) reducing the dynamic rupture velocity and stress drop. The experiments also show that 1) during the fault nucleation process, the b-value for AEs decreases from 1-1.3 to 0.5 before dynamic rupture, then rapidly recovers to around 1-1.2 during aftershock activity and 2) the hydraulic diffusivity gradually increases from an initial pre-rupture order of 0.1 m2/s to 10-100 m2/s after dynamic rupture. These results provide a reasonable fault pre-slip model, indicating that hydraulic fracturing promotes shear slip before dynamic rupture, as well as laboratory-scale insights into ensuring the safety and effectiveness of hydraulic fracturing operations related to activities such as geothermal development, evaluating the seismic risk induced by water injection, and further researching the precursory preparation process for deep fluid-driven or fluid-involved natural earthquakes. Potential uses of the data sets include but are not limited to 1) Providing training datasets for machine learning and AI-based technology development, such as developing machine learning models to predict stress accumulation and the remaining time before final fracture. 2) Developing effective methods for identifying weak or low S/N ratio AE signals. The waveform data contain numerous AE events that cannot be accurately located using conventional methods. 3) Inverting source mechanisms and moment tensors. Our research has not yet systematically analyzed the moment tensors of AE events. 4) Conducting more in-depth research on the interaction between fluid migration and rock deformation and fracture. Please cite the associated article as: X. Lei (2024), Fluid-driven fault nucleation, rupture processes, and permeability evolution in Oshima granite — Preliminary results and acoustic emission datasets, Geohazard Mechanics,



Hydraulic Rock Mechanics, Induced Seismicity