Sabine A. M. Den Hartog

Geological CO ₂ storage is transitioning from demonstration to industrial-scale deployment, and faults within and around storage complexes exert a first-order control on its safety and performance. Faults can compartmentalise reservoirs through across-fault sealing, provide pathways for vertical migration of CO ₂ and brine across caprocks, and host induced seismicity when reactivated by pore-pressure, thermal and geochemical changes during and after injection. These behaviours are not independent in the sense that the same fault can seal hydraulically while remaining critically stressed, or slip without becoming transmissive, and its state can evolve over engineering to geological timescales. We synthesise structural geology, fault-seal analysis, geomechanics, geochemistry, modelling and the operational record from projects including Sleipner, In Salah, Decatur, Quest, Snøhvit, Aquistore and Gorgon. No measurable fault-related leakage from a storage complex has been documented at any saline-aquifer or depleted-gas-field project. Microseismicity on unmapped sub-seismic and basement structures is recurrent, and in 2026 two events above M _L 4.0 near Quest became the first felt events at a storage project, provisionally attributed to injection and without effect on containment. Because the pressure and stress perturbation extends far beyond the CO ₂ plume, the faults that host seismicity commonly lie beneath or outside the storage complex and loading there can peak after injection stops. We review across-fault sealing proxies and their CO ₂ -specific uncertainty, fault-fracture network controls on leakage, induced-seismicity mechanisms, natural analogues, regulatory frameworks and post-injection behaviour, and identify five priority data classes and the coupled workflows needed to convert them into defensible risk envelopes.

The geochemical alteration of host rocks might affect the productivity and the potential for induced seismicity of geothermal systems. In addition to natural alteration, following production and heat extraction, re‐injected fluids at lower temperatures and different pressures may be in chemical disequilibrium with the rock, impacting mineral solubility and dissolution/precipitation processes. In this study, we investigate the effect of geochemical alteration on the frictional behavior of granites, and their seismogenic potential, by conducting direct shear experiments using samples with varying degrees of alteration. The samples originate from the Carnmenellis granite in Cornwall, SW England, and represent the formation used in the United Downs Deep Geothermal Power Project for heat extraction. Experiments were conducted on granite powders (referred to as gouges) at room temperature and 180°C, at simulated in situ confining and pore pressures of 130 and 50 MPa, respectively (∼5 km depth). With increasing degree of alteration, the frictional strength of the gouges decreases while frictional stability increases. At high temperature, frictional stability is reduced for all samples while maintaining the trend with alteration stage. Microstructural investigation of the sheared gouges shows alteration delocalizes shear by reducing grain size and increasing clay fraction, which promotes the formation of pervasive shear fabrics. Our work suggests that, within the range of tested pressures, more alteration of granite initially causes more stable shearing in a fault. This behavior with alteration is sustained at high temperatures, but the overall frictional stability is reduced which increases the potential for induced seismicity at higher temperatures.

Geochemical alteration of host rocks might affect the productivity and the potential for induced seismicity of geothermal systems. In addition to natural alteration, following production and heat extraction, re-injected fluids at lower temperatures and different pressures may be in chemical disequilibrium with the rock, impacting mineral solubility and dissolution / precipitation processes. In this study, we investigate the effect of geochemical alteration on the frictional behaviour of granites, and their seismogenic potential, by conducting direct shear experiments using samples with varying degrees of alteration. The samples originate from the Carnmenellis granite in Cornwall, SW England, and represent the formation used in the United Downs Deep Geothermal Power Project for heat extraction. Experiments were conducted on granite powders (referred to as gouges) at room temperature and 180°C, at simulated in-situ confining and pore pressures of 130 MPa and 50 MPa, respectively (~5 km depth). With increasing degree of alteration, the frictional strength of the gouges decreases while frictional stability increases. At high temperature, frictional stability is reduced for all samples while maintaining the trend with alteration stage. Microstructural investigation of the sheared gouges shows alteration delocalises shear by reducing grain size and increasing clay fraction, which promotes the formation of pervasive shear fabrics. Our work suggests that, within the range of tested pressures, more alteration of granite initially causes more stable shearing in a fault. This behaviour with alteration is sustained at high temperatures, but the overall frictional stability is reduced which increases the potential for induced seismicity at higher temperatures.

Subduction megathrusts exhibit a spectrum of slip modes, including catastrophic earthquakes. Although the mechanical and frictional properties of materials sampled from subduction zones have been studied extensively, few datasets have been collected for compositions and at pressure and temperature conditions representative of those in situ. The Nankai subduction zone in southwest Japan is a well‐studied margin, and abundant data provide an opportunity to advance our understanding of fault and earthquake physics. Here, we use samples exhumed in the Shimanto and Sanbagawa Belts on Shikoku Island of southwest Japan that represent analogs for materials along the present‐day megathrust at depths of ∼5–>25 km, and we shear these at their peak in situ pressure‐temperature (P‐T) conditions. Effective normal stresses range from 28 to 192 MPa, and temperatures from 105°C to 470°C. We used pore fluid pressures of 45–240 MPa, corresponding to fluid overpressure ratios λ of 0.65 and 0.90. Slip velocities of 0.1–100 μm/s were used, in order to focus on the nucleation of instability and earthquakes. We found predominantly velocity‐strengthening (inherently stable) behavior under all conditions for λ = 0.65. For λ = 0.90, velocity‐weakening behavior was observed at 350°C, with velocity‐strengthening behavior at lower and higher temperatures. The rate/state frictional stability parameter (a‐b) increases with slip velocity at temperatures up to ∼200°C and remains constant or decreases with slip velocity at higher temperatures. Overall, our results demonstrate the potentially important roles of both temperature and slip velocity in controlling the distribution of stress and frictional rheology along subduction thrusts.

Laboratory measurements of constitutive frictional parameters are commonly inferred to explain the wide variety of slip behavior seen on natural faults. The extent to which these small‐scale measurements directly relate to fault slip behavior remains obscure. In this work, we compare laboratory‐determined frictional parameters on surface‐derived samples from along the length of the Longitudinal Valley Fault (LVF) in Taiwan with the observed slip behavior and with frictional parameters obtained geodetically. The LVF displays partially locked and creeping sections and a M_w 6.8 event in 2003 produced transient acceleration of slip in the adjacent creeping sections that can be used to determine the frictional parameters for direct comparison with the laboratory‐measured ones. We find that the laboratory‐measured friction parameters are markedly different for samples collected from the creeping and partially locked sections of the fault, the former showing lower friction coefficients and more positive values of the fault stability parameter (a−b). Moreover, values for the product of (a−b) and the effective normal stress, determined geodetically, relate very closely to those measured in the laboratory. Mineralogical and microstructural analyses of the fault gouges show that some mineralogically similar gouges produce distinctly different frictional behavior, and that this may be related to the presence and distribution of kaolinite. We conclude overall that upscaling of laboratory measurements of fault frictional properties appears to reflect well the large‐scale slip behavior of faults.

Fault slip is often localized in phyllosilicate‐rich fault gouges in a manner consistent with the relatively low friction coefficients measured for dry and especially wet phyllosilicates in laboratory experiments. However, the microphysics controlling these low friction coefficients remains unclear. Here, we propose a microphysical model, inspired by microstructural observations, for the prediction of the absolute value of the friction coefficient of pure dry and wet phyllosilicates. Experimentally produced phyllosilicate gouges suggest that shearing is controlled by sliding along (001) grain/platelet interfaces operating in series with removal of overlapping grain edge barriers by basal cleavage. We derive a model incorporating a subcritical crack propagation equation for the latter, constrained by subcritical crack growth data for muscovite. Model predictions for muscovite show similar trends regarding the effects of humidity and slip velocity on friction coefficient as do experiments at room temperature. The absolute value predicted for the friction coefficient is difficult to compare with experimental values, as it critically depends on atomic scale (001) sliding resistance, which is poorly constrained by available experimental data. Further discrepancies with experimental data can be explained by effects of varying grain size, grain aspect ratio, and porosity on the friction coefficient. While numerous qualitative explanations have been proposed previously for the low friction coefficient exhibited by phyllosilicates, especially in the presence of water, our study provides a new step toward a quantitative, physically based model.

The impact of long‐term CO₂‐brine‐rock interactions on the frictional properties of faults is one of the main concerns when ensuring safe geological CO₂ storage. Mineralogical changes may alter the frictional strength and seismogenic potential of pre‐existing faults bounding a storage complex. However, most of these reactions are too slow to be reproduced on laboratory timescales and can only be assessed using geochemical modeling. We combined modeling of CO₂‐charged formation water and fault gouges (1–1000 years residence time, i.e. 10–10⁶ pore volume flushes) with friction experiments on simulated fault gouges (T = 22–150°C; σ_n^eff = 50 MPa; P_f = 25 MPa; V = 0.2‐100 μm/s), having mineralogical compositions as predicted by the models. As an analogue for clay‐rich caprocks overlying potential CO₂ storage sites in Europe, we used the Opalinus claystone. Our experiments showed that, although significant mineralogical changes occurred, they did not significantly change the frictional behavior of faults. Instead, initial fault‐gouge mineralogy imposed a stronger control on clay‐rich fault behavior than the extent of CO₂‐brine‐rock interactions, even under chemical conditions allowing for significant reaction. We demonstrated that the impact of mineralogical changes due to CO₂‐brine‐rock interactions on the frictional behavior and seismogenic potential of faults could be assessed using our combination of geochemical modeling and friction experiments. Note that a complete understanding requires evaluation of additional effects, such as that of shear velocity, effective normal stress, and other fault characteristics (maturity, shear strain). © 2018 The Authors. Greenhouse Gases: Science and Technology published by Society of Chemical Industry and John Wiley & Sons, Ltd.