Rough Fault Research Articles

Adjoint-based inversion for stress and frictional parameters in earthquake modeling

We present an adjoint-based optimization method to invert for stress and frictional parameters used in earthquake modeling. The forward problem is linear elastodynamics with nonlinear rate-and-state frictional faults. The misfit functional quantifies the difference between simulated and measured particle displacements or velocities at receiver locations. The misfit may include windowing or filtering operators. We derive the corresponding adjoint problem, which is linear elasticity with linearized rate-and-state friction and, for forward problems involving fault normal stress changes, nonzero fault opening, with time-dependent coefficients derived from the forward solution. The gradient of the misfit is efficiently computed by convolving forward and adjoint variables on the fault. The method thus extends the framework of full-waveform inversion to include frictional faults with rate-and-state friction. In addition, we present a space-time dual-consistent discretization of a dynamic rupture problem with a rough fault in antiplane shear, using high-order accurate summation-by-parts finite differences in combination with explicit Runge–Kutta time integration. The dual consistency of the discretization ensures that the discrete adjoint-based gradient is the exact gradient of the discrete misfit functional as well as a consistent approximation of the continuous gradient. Our theoretical results are corroborated by inversions with synthetic data. We anticipate that adjoint-based inversion of seismic and/or geodetic data will be a powerful tool for studying earthquake source processes; it can also be used to interpret laboratory friction experiments.

A scaling relationship for the width of secondary deformation around strike-slip faults

Simple mechanical arguments suggest that slip along interlocked, rough faults, damages surrounding rocks. The same arguments require that the scale of secondary damage is proportional to the size of geometric irregularities along the main fault. This relationship could apply at all scales, but has, so far, been difficult to observe at the 10s to 100 s of km scales of large, natural faults, often because large-scale deformation is distributed across wide, complex plate-boundary fault systems, like the San Andreas Fault. The geometry and geology of another large-scale plate-boundary strike slip fault—the Queen Charlotte Fault (QCF)—is, in contrast, especially simple. Here, we show that observations of secondary deformation are well-aligned with predictions of stress variations caused by geometric irregularities along the QCF, suggesting a geometric relationship between primary fault geometry and secondary deformation. The analytic stress solution reveals that the highest stresses and highest likelihood of failure are confined to a zone of influence (ZOI) with a width quantified by ZOI=λ/2π, where λ is the wavelength of geometric variations along the main fault. This simple model is consistent with ∼100-km-scale observations along the QCF and can theoretically be used to predict the width of secondary deformation at all scales.

Energy Dissipation During Shear Along Experimental Rough Faults

AbstractThe energy budget and the interplay between stable friction evolution and dynamic stick‐slips are tested here under continuous slip along rough faults. We conducted 34 direct‐shear experiments coupled with precise roughness measurements on diabase and limestone fault samples. The faults broad roughness ranges from highly rough and interlocked fractured interfaces to smooth polished surfaces. The analysis focuses on two slip phases: (a) the evolution of the shear strength of rough sample under stable, cumulative displacement; and (b) the dynamic of unstable stick‐slip sliding. We found that the breakdown work during frictional strength evolution increases with roughness increase across multiple scales. The diabase samples are more sensitive to roughness increase than limestone samples in terms of the breakdown work implied by frictional evolution. We attribute this increased diabase sensitivity with fault roughness to its higher bulk elasticity and not to the fault shear stiffness. The diabase faults displayed multiple periodic system‐size stick‐slips, and the measured stored energy during the preparatory stage were surprisingly independent of the fault roughness. This finding suggests that during the preparatory stage a balance between the intracycle fault stiffness and stress drop govern the stored energy magnitude. Further, this energy balance suggests that some interface conditioning occurs before the spontaneous slip overcomes a sticking barrier.

Statistical distribution of static stress resolved onto geometrically-rough faults

The in-situ stress state within fault zones is technically challenging to characterize, requiring the use of indirect methods to estimate. Most work to date has focused on understanding average properties of resolved stress on faults, but fault non-planarity should induce spatial variations in resolved static stress on a single fault. Assuming a particular stochastic model for fault geometry (band-limited fractal) gives an approximate analytic solution for the probability density function (PDF) on fault stress that depends on the mean fault orientation, mean stress ratio, and roughness level. The mean stress is shown to be equal to the planar fault value, while deviations are described by substituting a second-order polynomial expansion of the stress ratio into the inverse distribution on fault slope. The result is an analytical expression for the PDF of shear-to-normal stress ratio on 2-D rough faults in a uniform background stress field. Two end-member distributions exist, one approximately Gaussian when all points on the fault are well away from failure, and one reverse exponential, which occurs when the mean stress ratio approaches the peak. For the range of roughness values expected to apply to crustal faults, stress deviations due to geometry can reach nearly 100% of the background stress level. Consideration of such a distribution of stress on faults suggests that geometric roughness and the resulting stress deviations may play a key role in controlling earthquake behavior.

Empowering Machine Learning Forecasting of Labquake Using Event‐Based Features and Clustering Characteristics

AbstractFollowing recent advances of machine learning (ML), we present a novel approach to extract spatiotemporal seismo‐mechanical features from Acoustic Emission (AE) catalogs to empower ML‐based forecasting. The AE data were recorded during laboratory stick‐slip experiments on granite samples cut by rough faults. Based on the features computed for a past time window, a random forest (RF) classifier is used to forecast the occurrence of a large magnitude event (MAE > 3.5) in the next time window. Event‐based features allow us to associate informative time‐space characteristics to each feature and nearest‐neighbor clustering analysis enables us to separate background and clustered seismicity and train individual models. The results show that the separation of AEs enhances the forecasting accuracy from 73.2% for the entire catalog up to 82.1% and 89.0% if background and clustered events are used separately. The presented new approach may be upscaled for applications to forecast tectonic earthquakes.

The evolution of stresses and shear resistance on rough faults at large slip

The deviation of natural faults from planarity results in geometric asperities and a heterogeneous stress field, affecting the shear resistance and slip behavior. This study numerically examines the effects of macro-scale roughness, local friction, and wear on the evolution of the fault stresses and shear resistance with fault slip. In contrast to previous analytical and numerical studies, our quasistatic simulations allow for slip at the scale of roughness wavelengths, large roughness levels, and fault opening. The simulations indicate that analytical solutions, which predict a linear increase in shear resistance with slip, significantly overestimate the additional shear resistance from roughness, i.e., the roughness drag, at slips larger than several percent of the minimum roughness wavelength. Starting with mated fault surfaces, the average shear resistance on the fault initially rises rapidly, but the rate of increase diminishes significantly as slip increases and the faults open. While initially showing a larger increase in shear resistance than self-similar faults, at large slip, self-affine faults exhibit significantly smaller shear resistance, and the background stresses are bounded without the need for off-fault inelastic deformation. Although smaller than predicted by analytical solutions, the simulation results show that the additional resistance from macro-scale geometrical irregularities is important, enabling rough faults to operate under shear resistance significantly larger than that from local friction only. The impact of macro-scale roughness increases with the roughness amplitude and the friction coefficient, while wear reduces fault stresses and average shear resistance. The combined effect of local friction and roughness is nonlinear, involving a direct contribution to shear resistance from roughness, as observed in prior studies, and an additional contribution from the increasing average normal stress with slip, which amplifies the resistance from friction and becomes more pronounced at large slip.

The different effects of polished and post-slip roughnesses on fault stability

Linking fault morphology to its mechanical properties in laboratory experiments may shed some light on analyzing the seismogenic potential of a natural fault. Many previous experiments suggested that increasing the roughness of a polished fault tends to stabilize the fault. Furthermore, the roughness of faults, both experimental and natural, spontaneously evolves during slip. Here, we investigate the effects of the polished roughness and the post-slip roughness on experimental fault stability. The experiments were conducted on a granodiorite fault under initial normal stresses of 5–6 MPa and total slip distances of 1.5–2.5 mm. The results reveal that the polished fault with a higher roughness exhibits stick-slip, while the post-slip fault with a lower roughness exhibits stable slip. Moreover, a decrease in the roughness of a post-slip fault tends to stabilize the fault. The monitored macroscopic loading data indicate that the critical slip distance is significantly greater in a post-slip fault than in a polished fault. These results imply that the critical slip distance of young, rough faults may evolve dramatically with slip, subsequently affecting the fault stability.

Mesh size effect on finite source inversion with 3-D finite-element modelling

SUMMARY Three-dimensional finite-element models, which can handle the stress perturbations caused by subsurface mechanical heterogeneities and fault interactions, have been combined with the finite source inversion to estimate the coseismic slip distribution over the fault plane. However, the mesh grid for discretizing the governing equations in the finite-element model significantly affects the numerical accuracy. In this study, we performed kinematic finite source inversion with idealized (regular observation point array; M1A–M1D) and regional (GEONET, GPS Earthquake Observation Network System stations in Japan; M2A–M2H) models with different mesh sizes to quantitatively analyse the effect of the mesh grid size around the fault plane on the inverted fault slip distribution. Synthetic observation data vectors obtained from the finest models (M1A and M2A) are compared with those from the coarser models (M1B–M1D and M2B–M2H), which were adopted to construct Green's function matrix. We found that the coarser mesh models derived a smaller surface displacement, leading to a decrease in the norm of Green's function matrix, which in turn influences the fault slip magnitude from the finite source inversion. Finally, we performed the source inversion for the fault slip distribution of the 2011 Mw 9.0 Tohoku–Oki earthquake using the coseismic surface displacements recorded at the GEONET and seafloor stations and finite-element modelling. By reducing the mesh size on the fault, we confirmed that the estimated magnitude of fault slip converged to approximately 80 m, which is consistent with the range of fault slip amounts from previous studies based on the Okada model. At least 0.88 million total domain elements and a 6.7 km2 mesh size on the fault plane with an area of 240 × 720 km2 are required for the convergence of the fault slip. Furthermore, we found that the location of the maximum fault slip is less sensitive to the mesh size, implying that source inversion based on a coarse mesh model (i.e. less than 0.5 million elements and > ∼60 km2 mesh size) can quickly provide the rough fault slip distribution.

Precursory Signals to Injection Induced Fault Reactivation in the Laboratory Using Active Ultrasonic Monitoring Methods

AbstractInduced earthquakes are still highly unpredictable, and often caused by variations in pore fluid pressure. Monitoring and understanding the mechanisms of fluid‐induced fault slip is essential for seismic risk mitigation and seismicity forecasting. Fluid‐induced slip experiments were performed on critically stressed faulted sandstone samples, and the evolution of the actively sent ultrasonic waves throughout the experiment was measured. Two different fault types were used: smooth saw‐cut fault samples at a 35° angle, and a rough fault created by in situ faulting of the samples. Variations in the seismic slip velocity and friction along the fault plane were identified by the coda of the ultrasonic waves. Additionally, ultrasonic amplitudes show precursory signals to laboratory fault reactivation. Our results show that small and local variations in stress before fault failure can be inferred using coda wave interferometry for time‐lapse monitoring, as coda waves are more sensitive to small perturbations in a medium than direct waves. Hence, these signals can be used as precursors to laboratory fault slip and to give insight into reactivation mechanisms. Our results show that time‐lapse monitoring of coda waves can be used to monitor local stress changes associated with fault reactivation in this laboratory setting of fluid‐induced fault reactivation. This is a critical first step toward a method for continuous monitoring of natural fault zones, contributing to seismic risk mitigation of induced and natural earthquakes.

Fault roughness controls injection-induced seismicity

Surface roughness ubiquitously prevails in natural faults across various length scales. Despite extensive studies highlighting the important role of fault geometry in the dynamics of tectonic earthquakes, whether and how fault roughness affects fluid-induced seismicity remains elusive. Here, we investigate the effects of fault geometry and stress heterogeneity on fluid-induced fault slip and associated seismicity characteristics using laboratory experiments and numerical modeling. We perform fluid injection experiments on quartz-rich sandstone samples containing either a smooth or a rough fault. We find that geometrical roughness slows down injection-induced fault slip and reduces macroscopic slip velocities and fault slip-weakening rates. Stress heterogeneity and roughness control hypocenter distribution, frequency-magnitude characteristics, and source mechanisms of injection-induced acoustic emissions (AEs) (analogous to natural seismicity). In contrast to smooth faults where injection-induced AEs are uniformly distributed, slip on rough faults produces spatially localized AEs with pronounced non-double-couple source mechanisms. We demonstrate that these clustered AEs occur around highly stressed asperities where induced local slip rates are higher, accompanied by lower Gutenberg-Richter b-values. Our findings suggest that real-time monitoring of induced microseismicity during fluid injection may allow identifying progressive localization of seismic activity and improve forecasting of runaway events.

Injection-induced fault slip and associated seismicity in the lab: Insights from source mechanisms, local stress states and fault geometry

Probing source mechanisms of natural and induced earthquakes is a powerful tool to unveil associated rupture kinematics. The source processes of failure and slip instability driven by stress loading are affected by fault geometry, but the source ruptures of injection-induced seismicity in relation to fault structures and local stress states remain poorly understood. We have conducted a series of fault reactivation and slip experiments on sandstone samples containing faults with different surface roughness (smooth saw-cut fault and fractured rough fault). We impose progressive fluid injection to induce fault slip, and simultaneously monitor the associated acoustic emission (AE) activity. Using high-resolution AE recordings, we perform full moment tensor inversion of all located AE sources, and investigate the changes of AE source characteristics associated with induced fault slip and their relation to fault roughness. For the complex and rough fault, we observe significant non-double-couple components of AE sources and a high degree of focal mechanism heterogeneity. The temporal changes of AE mechanisms associated with injection-induced fault slip on the smooth fault reveal increasing proportions of double-couple components and decreasing variability of AE focal mechanisms when approaching the onset of slip events. The observed inconsistency between the nodal planes of AE sources and the macroscopic fault plane orientation is attributed to the development of secondary fracture networks surrounding the principal slip surface. We analyze changes in the magnitude-frequency characteristics and source mechanisms of AEs with fault-normal distance, showing that for the smooth (mature) fault, Gutenberg–Richter b-value of on-fault seismicity is lower and focal mechanisms are less heterogeneous, compared to off-fault seismicity. Our results emphasize the important role of roughness-related changes in local fault geometry and associated stress heterogeneity for source mechanisms and rupture kinematics of injection-induced seismicity.

Study of the Migrating Mine Gas Piston Effect during Reactivation of Tectonic Faults

The hypothesis of the piston effect during mine gas migration caused by fault reactivation was studied, with the use of computer modeling, to explain cases of a sudden appearance of mine gases on the earth surface in coal mining areas. The study is based on the factual data of the mode and amplitudes of subsidence along faults during mining, the morphology of the fault planes, and the theoretical ideas about the discrete nature of the fault wall displacement along uneven contact surfaces. It is taken into account that the walls of the fault, due to the asperity types “ridge” and “sag”, form contacts of the “ridge–ridge” and “ridge–sag” patterns. This study examines the situation where gas pressure in the fracture space can sharply increase due to the jerky displacement of reactivated tectonic fault walls with a rough fault plane. It is assumed that, in the first phase of reactivation, the fracture space expands as a result of the displacer opening and the fact that fault plane asperities engage in the “ridge–ridge” type of contact. With the subsequent relative displacement of the fault walls in the second phase of reactivation, the contact changes into the “ridge–sag” type and a sharp reduction in the fracture space volume occurs. It is shown that a “piston effect” emerges due to the reduction in fracture space and that it promotes an increase in gas pressure and stimulates gas movement to the surface through the available channels. The resulting “piston effects” may also be responsible for the suddenly raised gas content recorded in the air of surface structures and recurrent mine gas migration onto the surface. The findings expand our understanding of the sudden gasification of the earth surface and living spaces in coal mining areas and contribute to the understanding of the gas migration process, thereby helping to monitor hazards.

Explainable machine learning for labquake prediction using catalog-driven features

Recently, Machine learning (ML) has been widely utilized for laboratory earthquake (labquake) prediction using various types of data. This study pioneers in time to failure (TTF) prediction based on ML using acoustic emission (AE) records from three laboratory stick-slip experiments performed on Westerly granite samples with naturally fractured rough faults, more similar to the heterogeneous fault structures in the nature. 47 catalog-driven seismo-mechanical and statistical features are extracted introducing some new features based on focal mechanism. A regression voting ensemble of Long-Short Term Memory (LSTM) networks predicts TTF with a coefficient of determination (R2) of 70% on the test dataset. Feature importance analysis revealed that AE rate, correlation integral, event proximity, and focal mechanism-based features are the most important features for TTF prediction. Results reveal that the network uses all information among the features for prediction, including general trends in high correlated features as well as fine details about local variations and fault evolution involved in low correlated features. Therefore, some highly correlated and physically meaningful features may be considered less important for TTF prediction due to their correlation with other important features. Our study provides a ground for applying catalog-driven to constrain TTF of complex heterogeneous rough faults, which is capable to be developed for real application.

Preparatory Slip in Laboratory Faults: Effects of Roughness and Load Point Velocity

AbstractAseismic slip may occur during a long preparatory phase preceding earthquakes, and what controls it remains poorly understood. In this study, we explored the role of load point velocity and surface roughness on slow slip during the preparatory stage prior to stick‐slip events. To that end, we conducted displacement‐rate controlled friction experiments by imposing varying load point velocities on sawcut granite samples with different surface roughness at a confining pressure of 35 MPa. We measured the average slip along the fault with the recorded far‐field displacements and strain changes, while acoustic emission sensors and local strain gages were used to capture local slip variations. We found that the average amount of aseismic slip during the preparatory stage increases with roughness, whereas precursory slip duration decreases with increased load point velocity. These results reveal a complex slip pattern on rough faults which leads to dynamic ruptures at high load point velocities.

Modeling earthquake cycles and wear on rough faults with the mortar finite element method

SUMMARYThis paper presents a mortar-based finite element formulation for modelling earthquake cycles and wear on rough faults governed by rate and state friction. The method allows large sliding on the fault and accounts for all stages in the earthquake cycle, using a variable time step size with a transition between quasi-static and fully dynamic time discretizations. Wear laws with linear and power law forms are discretized and implemented into the mortar method, as well as a minimum level of normal traction constraint to treat fault opening. We examine the effect of wear laws on the slip behaviour and near-fault stresses during simulations of earthquake cycles on rough faults. The simulations demonstrate that the implementation of wear allows more realistic modelling of the earthquake cycle without the development of unrealistically large stresses and with less reduction of earthquake magnitude with total fault slip. Moreover, the method enables to study the effects of roughness and fault slip on the gouge zone thickness.

Method to quantify the dynamic near-fault full-field stress evolution associated with rough fault shear deformation based on 3D printed models and photoelastic measurements

Quantifying the dynamic near-fault stress evolution associated with fault shear deformation is crucial for revealing the earthquake nucleation mechanisms of rough faults. However, it is still challenging via traditional methods because of the difficulties in resolving continuously evolved stresses and high stress heterogeneity. In this study, we present a novel method to quantify the dynamic full-field stress evolution using three-dimensional printed models and photoelastic techniques. The dynamic full-field stresses in a circular disc determined by this method and an analytical model are compared to verify the accuracy of the proposed method. A new method to count the fringe orders near the rough fault is proposed to achieve the stress quantification. The full-field principal stress difference in a rough fault model under shear forces is derived. Our findings indicate that the proposed method can significantly improve the efficiency of stress calculation and effectively quantify the dynamic full-field stress near rough faults.

Dynamic simulations of coseismic slickenlines on non-planar and rough faults

SUMMARY Knowing the directions of rupture propagation of palaeo-earthquakes is a challenging, yet important task for our understanding of earthquake physics and seismic hazard, as the rupture direction significantly influences the distribution of strong ground motion. Recent studies proposed a relationship between the direction of rupture propagation and curvature of slickenlines formed during seismic slip. The relationship was established using a global catalogue of historic surface-rupturing earthquakes and dynamic models of idealized, planar faults. At the same time, some slickenlines previously documented on geometrically complex fault segments show their convexity opposite from the simple model prediction, which we refer to as ‘abnormal convexity’. To explain such observations, we perform simulations of spontaneous earthquake ruptures on non-planar and rough faults. We find that in the case of strike-slip faults, a non-planar fault model can lead to abnormal convexity of slickenlines in places where the fault dip angle changes abruptly from the average dip of the fault. Abnormal convexity of slickenlines is produced when the initial along-dip stresses are larger than, and are opposite in direction to, the dynamic stresses imparted by the mixed-mode rupture. Such results are also confirmed in our rupture simulations with a rough fault. Our results also show that the parameter space for which abnormal convexity of slickenlines occurs near Earth’s surface is narrow, especially when the fault strength and initial shear stresses increase with depth. Nevertheless, slickenlines on geometrically complex faults need to be carefully interpreted and investigation of rupture direction using curved slickenlines should focus on structurally simple parts of faults.

Frictional behavior and micro-damage characteristics of rough granite fractures

Natural faults or discontinuities in general are usually rough at different scales, and the roughness of fault surfaces plays a dominant role in determining the mode of sliding (stable or unstable). Understanding the frictional behavior of rough fractures is important for investigating the mechanics and nucleation of earthquakes, and other dynamic geo-hazards such as induced seismicity and fault slip rockburst. We perform direct shear tests on rough granite fractures to investigate the shear behavior, microscopic asperity damage and surface wear characteristics under different normal stresses (10 to 40 MPa). We then compare our test results with those obtained using ground saw-cut fault surfaces. Results indicate that stick-slip occurs on all the rough fractures during the sliding stage, and the amount of stress drops and fault gouges tends to increase with normal stress. The peak friction coefficient is much higher than that of ground fault surfaces, while the steady-state friction coefficient is comparable with that of the ground surfaces. We identify three different scales of brittle fractures, which are considered to be closely related to the stick-slip sliding, namely shear-off of asperities, fracturing of the survived asperities and off-fault tensile fractures propagating into the host rock. The micro-damage of the post-shear surface occurs in the form of powder-sized gouge, micro-cracks and grain edge wear; and the micro-crack number, maximum length and aperture tend to increase with increasing normal stress. Influence of different fault types (rough fractures, ground saw-cut surfaces and gouges) on the stress drop is not very discernable, which is partly because that stick-slip is highly susceptible to different loading conditions. Friction drops for different fault types are found to be clustered around 0.1 but lower than 0.3 below 50 MPa normal stress, which also depend on the loading conditons (e.g., stiffness). Our study contributes to a better understanding of the mechanics underlying the dynamic geohazards (earthquakes, induced seismicity, fault slip rockburst) which are associated with the unstable shear failure on rough faults or fracture surfaces.

Precursory Off‐Fault Deformation in Restraining and Releasing Step Overs: Insights From Discrete Element Method Models

AbstractAccelerating geophysical activity is detected preceding some, but not all, large earthquakes. This observation may indicate that no precursors occur before some earthquakes, or that the instrumentation lacks the required sensitivity. To aid crustal monitoring efforts, we use discrete element method models to identify the locations and styles of deformation that may provide useful information about approaching fault reactivation. We model the reactivation of two healed rough faults in a variety of step over configurations, embedded in a host rock with varying amounts of damage subject to shear velocity loading parallel to the faults. Both the fault geometry and ratio of fault to host rock strength control the amount of off‐fault deformation. Consistent with field observations, models with larger steps and more preexisting host rock damage produce higher amounts of off‐fault deformation. We assess the size of the continuous regions of high velocities and strains to compare the value of the precursory information of each velocity and strain component. Comparing the three components of the velocity vector suggests that the fault‐parallel velocity produces the largest and most temporally continuous regions of elevated velocity. The size of these regions increases toward failure, indicating the usefulness of tracking this component. Comparing the volumetric and shear components of the three‐dimensional strain tensor suggests that during most of the interseismic period, the shear strain provides more information about approaching fault slip than the volumetric strain. However, in the days and months preceding fault reactivation, both the shear and volumetric strains provide similarly valuable information.

The Influence of Roughness on Experimental Fault Mechanical Behavior and Associated Microseismicity.

Fault surfaces are rough at all scales, and this significantly affects fault‐slip behavior. However, roughness is only occasionally considered experimentally and then often in experiments imposing a low‐slip velocity, corresponding to the initiation stage of the earthquake cycle. Here, the effect of roughness on earthquake nucleation up to runaway slip is investigated through a series of dry load‐stepping biaxial experiments performed on bare rock surfaces with a variety of roughnesses. These laboratory faults reached slip velocities of at least 100 mm/s. Acoustic emissions were located during deformation on bare rock surfaces in a biaxial apparatus during load‐stepping experiments for the first time. Smooth surfaces showed more frequent slip instabilities accompanied by slip bursts and larger stress drops than rough faults. Smooth surfaces reached higher slip velocities and were less inclined to display velocity‐strengthening behavior. The recorded and localized acoustic emissions were characterized by a greater proportion of large‐magnitude events, and therefore likely a higher Gutenberg‐Richter b GR‐value, for smoother samples, while the cumulative seismic moment was similar for all roughnesses. These experiments shed light on how local microscopic heterogeneity associated with surface topography can influence the macroscopic stability of frictional interfaces and the associated microseismicity. They further provide a laboratory demonstration of roughness' ability to induce stress barriers, which can halt rupture, a phenomenon previously shown numerically.