In this study, a high-confining pressure and real-time large-displacement shearing-flow setup was developed. The test setup can be used to analyze the injection pressure conditions that increase the hydro-shearing permeability and injection-induced seismicity during hot dry rock geothermal extraction. For optimizing injection strategies and improving engineering safety, real-time permeability, deformation, and energy release characteristics of fractured granite samples driven by injected water pressure under different critical sliding conditions were evaluated. The results indicated that: (1) A low injection water pressure induced intermittent small-deformation stick-slip behavior in fractures, and a high injection pressure primarily caused continuous high-speed large-deformation sliding in fractures. The optimal injection water pressure range was defined for enhancing hydraulic shear permeability and preventing large injection-induced earthquakes. (2) Under the same experimental conditions, fracture sliding was deemed as the major factor that enhanced the hydraulic shear-permeability enhancement and the maximum permeability increased by 36.54 and 41.59 times, respectively, in above two slip modes. (3) Based on the real-time transient evolution of water pressure during fracture sliding, the variation coefficients of slip rate, permeability, and water pressure were fitted, and the results were different from those measured under quasi-static conditions. (4) The maximum and minimum shear strength criteria for injection-induced fracture sliding were also determined (m = 0.6665 and m = 0.1645, respectively, m is friction coefficient). Using the 3D (three-dimensional) fracture surface scanning technology, the weakening effect of injection pressure on fracture surface damage characteristics was determined, which provided evidence for the geological markers of fault sliding mode and sliding nature transitions under the fluid influence. (c) 2025 Institute of Rock and Soil Mechanics, Chinese Academy of Sciences. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/ 4.0/).

Injection-induced seismicity has been a focus of industry for decades as it poses great challenges to the associated risk mitigation and hazard assessment. The response surface methodology is integrated into the geo-mechanical model to analyze the effects of multiple factors on induced seismicity during the post shut-in period. We investigate the roles of poroelastic stress and pore pressure diffusion and examine the differences in the controlling mechanism between fault damage zones and the fault core. A sensitivity analysis is conducted to rank the selected factors, followed by a Box-Behnken design to form response surfaces and formulate prediction models for the Coulomb stress and its components. Reservoir properties significantly affect the potentials of induced seismicity in the fault by changing pore pressure diffusion, which can be influenced by other factors to varying degrees. Coulomb stress is greater in pressurized damage zones than in fault cores, and the seismicity rate exhibits a consistent variation. Poroelastic stress plays a similar role to pore pressure diffusion in the stability of the fault within the pressurized damage zones. However, pore pressure diffusion dominates in the fault core due to the low rigidity, which limits the accumulation of elastic energy caused by poroelastic coupling. The slip along the fault core is a critical issue to consider. The potential for induced seismicity is reduced in the right damage zones as the pore pressure diffusion is blocked by the low-permeability fault core. However, poroelastic stressing still occurs, and in deep basements, the poroelastic effect is dominant even without a direct increase in pore pressure. The findings in this study reveal the fundamental mechanisms behind injection-induced seismicity and provide guidance for optimizing injection schemes in specific situations. (c) 2024 Institute of Rock and Soil Mechanics, Chinese Academy of Sciences. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/ 4.0/).