Description: Earthquakes associated with fluid injection in various geo-energy settings, such as shale gas and deep geothermal energy, have shelved many projects with great potential. However, the injection-rate dependence of earthquake nucleation length, i.e., the slowly slipping (creeping) fault length in preparation for a subsequent earthquake (Kaneko & Lapusta, 2008), remains elusive. In this study, we take a step towards this issue by performing fluid injection experiments on low-permeability granite samples containing a critically stressed sawcut fault at different local injection rates (0.2 mL/min and 0.8 mL/min) and confining pressures (31 MPa and 61 MPa) (c. f., Ji & Wu, 2017; Wang et al., 2020). An array of local strain gauges and acoustic emission (AE) hypocenter locations were used to monitor the precursory slip of critically stressed faults before injection-induced stick-slip failure (c. f., Passelègue et al., 2020; Wang et al., 2020). The nucleation length was determined for each injection-induced stick-slip event, and its dependence on effective normal stress and injection rate was explored. Herein, we compile the processed data obtained from the experiments in four Excel worksheets. The full description of the methods is provided in Ji et al. (2022).

Description: The utilisation of geothermal energy in the Netherlands is primarily focused on deep sedimentary aquifers, which are often intersected by major faults. Geothermal operations (i.e. fluid production and injection) may alter the effective stress state along these faults and trigger induced seismic events. Pore pressure perturbations have been generally considered the main driver of injection-induced seismicity. However, thermal stresses caused by temperature gradients between the re-injected cold fluid and the reservoir rock may also contribute to the triggering of earthquakes in geothermal reservoirs. While existing geothermal power plants operating in sandstone reservoirs did not produce any major induced seismicity, it is a matter of debate whether a reduction in the temperature of the re-injected fluid could increase the seismic hazard potential. In this study, we applied modified Gutenberg–Richter statistics based on frictional Coulomb stress variations implemented in a coupled thermo-hydro-mechanical model to estimate the seismic hazard caused by the operation of a geothermal doublet. We conducted a systematic parametric study to assess and rank the impact of different intrinsic (geological) and extrinsic (operational) parameters on the induced seismic hazard potential. We identified a competing mechanism between induced variations in pore pressure and thermal stress within the reservoir in controlling induced seismicity. We found that stress changes induced by pore pressure variations are the main cause of seismic hazard, although thermally induced stresses also contribute significantly. The results indicate that by optimising the operational parameters it is possible to increase production efficiency while maintaining a long-term control over the fluid injection-induced seismicity.

Description: Post shut‐in seismic events in enhanced geothermal systems (EGSs) occur predominantly at the outer rim of the co‐injection seismic cloud. The concept of postinjection fracture and fault closure near the injection well has been proposed and validated as a mechanism for enhancing post shut‐in pressure diffusion that promotes seismic hazard. This phenomenon is primarily attributed to the poroelastic closure of fractures resulting from the reduction of wellbore pressure after injection termination. However, the thermal effects in EGSs, mainly including heat transfer and thermal stress, may not be trivial and their role in postinjection fault closure and pressure evolution needs to be explored. In this study, we performed numerical simulations to analyze the relative importance of poroelasticity, heat transfer, and thermo‐elasticity in promoting postinjection fault closure and pressure diffusion. The numerical model was first validated against analytical solutions in terms of fluid pressure diffusion and against heated flowthrough experiments in terms of thermal processes. We then quantified and distinguished the contribution of each individual mechanism by comparing four different shut‐in scenarios simulated under different coupled conditions. Our results highlight the importance of poro‐elastic fault closure in promoting postinjection pressure buildup and seismicity, and suggest that heat transfer can further augment the fault closure‐induced pressure increase and thus potentially intensify the postinjection seismic hazard, with minimal contribution from thermo‐elasticity.

Description: Deep geothermal energy (~3-6 km depth) is a candidate for sustainable and carbonfree energy supply. One of the main concerns of deep geothermal systems is induced seismicity that may produce earthquakes of economic concerns, challenging the development of this form of alternative energy. So far, cold water injection has been overlooked but may contribute to induced seismicity due to fault reactivation through thermal stresses also beyond the cooling region. This can be of importance, in particular, in fractured and faulted geothermal reservoirs. In this study, we first compare different approaches to estimate induced seismic risk from slip-tendency analysis, rate-andstate friction theory and modified Gutenberg-Richter statistics based on frictional Coulomb-stress perturbations. Then, we systematically investigate effects of both, intrinsic geological parameters (e.g., fault-, host rock properties and in-situ stress), and operational parameters (e.g., well geometry and placement, injection schemes, induced pressure perturbation) on induced seismicity.

Description: Fault zones often serve as the major fluid pathways in a variety of geo-energy systems, such as deep geothermal systems. However, injection-induced instability of faults can sometimes lead to large-magnitude earthquakes. Cyclic injection has thus been proposed as an alternative injection protocol to better manage and mitigate the associated seismic risks. The risks of injection-induced seismicity depend primarily on the extent and magnitude of the fluid pressure perturbation. When fluid is injected into a fault zone, the local fault permeability will be enhanced, which in turn promotes the migration of fluid along the fault. This nonlinear process is further complicated during cyclic injection via alternating the injection pressure. In this study, both numerical and analytical modeling are conducted to investigate cyclic fluid injection into a fault zone with pressure sensitive permeability, in which the local fault permeability changes as a function of the local effective stress. The match with laboratory-scale experimental and field-scale analytical results of cyclic fluid injection verifies the accuracy of the numerical model. The parametric study reveals that the injection pressure attenuation, quantified by the amplitude ratio and phase shift, is enhanced by a lower initial fault permeability, a smaller stress sensitivity coefficient, and a shorter period of pressure cycle (i.e. a higher frequency). Besides, the amplitude of the pressure cycle has a negligible effect on the injection pressure attenuation. We also discuss the implications of our results for the less amenable far-field seismic hazard and post shut-in seismicity.

Description: Induced seismicity associated with fluid injection has raised serious concerns for the safety and efficiency of geo-energy systems. Cyclic injection has recently been proposed as an alternative injection scheme to reduce the large magnitude injection-induced seismicity. However, the influence of cyclic injection on the activation of natural fractures in granite and the resulting seismic risk is not yet clear. This study investigates the injection-induced activation of a critically stressed natural fracture in a granite core sample, particularly focusing on the comparison between monotonic and cyclic water injection under pressure-controlled and volume-controlled conditions. Experimental results show that the acceleration and deceleration of fracture slip are modulated by the shear stress imbalance between the fixed shear stress and the evolving frictional strength of the fracture. Fracture slip affects the fluid pressure distribution on the fracture, which in turn regulates the frictional strength of the fracture. At a small total shear displacement (i.e., ~ 0.9 mm in this study), cyclic injection with a restricted peak injection pressure results in aseismic fracture slip at much smaller peak slip rates compared to that during the monotonic injection. On the one hand, the more uniform reduction in effective normal stress caused by cyclic injection encourages slow and stable fracture slip, characterized by the smaller peak slip rates. On the other hand, the flowback of injected fluid or suspension of injection could prevent the occurrence of fast-accelerated fracture slip during cyclic injection. However, the fracture can become unstable when it has experienced a considerable amount of total shear displacement (larger than ~ 0.9 mm in this study), and likely gained a significantly enhanced permeability. Continued injection after the unstable shut-in stage, signified by an unusual increase in slip rate and an accelerated drop in injection pressure, could result in rapid and unstable fracture slip.

Description: Laboratory hydraulic fracturing tests on cubic granite specimens with a side length of 100 mm were performed under true triaxial stress conditions combined with acoustic emission monitoring. Six different injection schemes were applied to investigate the influence of the injection scheme on hydraulic performance and induced seismicity during hydraulic fracturing. Three of these schemes are injection rate controlled: constant rate continuous injection (CCI), stepwise rate continuous injection (SCI), and cyclic progressive injection (CPI); the other three are pressurization rate controlled: stepwise pressurization (SP), stepwise pulse pressurization (SPP) and cyclic pulse pressurization (CPP). The test results show that the SPP scheme achieves the highest increase in injectivity among the six schemes. The CPI scheme generates the lowest induced seismicity while the improvement in injectivity is the least pronounced. The CPP scheme allows increasing injectivity and decreasing induced seismicity, and is suggested as a promising alternative injection scheme for field applications. Thin section microscopic observations of fractured specimens show that intragranular fractures splitting microcline, orthoclase and quartz grains dominate the hydraulic fractures independent of the injection scheme. The SPP scheme creates the largest fracture length, which explains the highest injectivity value among all schemes. Tests with relatively low magnitude of maximum AE amplitude correspond to short fracture length and small portions of intragranular fractures in microcline grains. Quartz grains are more fractured than microcline and orthoclase grains, and quartz chips (natural proppants) are frequently observed adjacent to hydraulic fractures. The laboratory test results show the potential for hydraulic fracture growth control in field applications by advanced fluid injection schemes, i.e. cyclic pulse pressurization of granitic rock mass.

Description: Geothermal energy is one of the most viable sources of renewable heat. However, the potential risk of induced seismicity associated with geothermal operations may slow down the growth of the geothermal sector. Previous research has led to significant progress in understanding fluidinjection- induced seismicity in geothermal reservoirs. However, an in-depth assessment of thermal effects on the seismic risk was generally considered to be of secondary importance. This study aims to investigate the relative influence of temperature and key geological and operational parameters on the slip tendency of pre-existing faults. This is done through coupled thermo-hydro-mechanical simulations of the injection and production processes in synthetic geothermal reservoir models of the most utilized and potentially exploitable Dutch geothermal reservoir formations: Slochteren sandstone, Delft sandstone and Dinantian limestone. In our study, changes in the slip tendency of a fault can largely be attributed to thermo-elastic effects, which confirms the findings of recent studies linking thermal stresses to induced seismicity. While the direct pore pressure effect on slip tendency tends to dominate over the early phase of the operations, once pore pressure equilibrium is established in a doublet system, it is the additional stress change associated with the growing cold-water front around the injection well that has the greatest influence. Therefore, the most significant increase in the slip tendency was observed when this low-temperature front reached the fault zone. The distance between an injection well and a pre-existing fault thus plays a pivotal role in determining the mechanical stability of a fault. A careful selection of a suitable target formation together with an appropriate planning of the operational parameters is also crucial to mitigate the risk of induced seismicity. Besides the well-known relevance of the in situ stress field and local fault geometry, rock-mechanical properties and operation conditions exert a major influence on induced stress changes and therefore on the fault (re)activation potential during geothermal operations.

Description: Earlier experiments have shown that cyclic hydraulic fracturing (CHF) systematically reduces the monotonic breakdown pressure (MBP). However, cyclic injection also causes a significantly longer injection time to failure as compared to the monotonic injection tests and complex fracture propagation that is hard to predict. In this study, a different injection scheme employing rock fatigue behavior, named creep injection, was tested on granite cylinders. The creep injection creates continuous pressurization under a constant borehole pressure (CBP) with a pre-defined maximum value below the MBP. Three different pressure ratios (CBP/MBP) of 0.85, 0.9 and 0.95 were tested. We found that both the CHF and hydraulic fracturing with creep injection can reduce the breakdown pressure by ca. 15 ~ 20% without confining pressure. Two mechanisms could explain the reduction: the influence of fluid infiltration within the theory of linear poroelasticity and stress corrosion within the subcritical crack growth theory. The lifetime of the granite cores subjected to creep injection is comparable with previous CHF experiments employing the same pressure ratio. In addition, the lifetime increases logarithmically when the ratio of CBP/MBP is decreased. This relationship has a high regression coefficient of R2 = 0.97, and the lifetime can be well predicted using a stress corrosion index of 70. On the contrary, CHF shows a significantly larger variance in the lifetime with a regression coefficient of R2 = 0.19 and, therefore, is hard to predict. Our results also point out that the injection scheme can modify hydraulic fracture patterns, in terms of fracture aperture, branching, and fracture propagation.