Orals: Tue, 5 May, 08:30–10:15 | Room G2
The faulting caused by dyke intrusions provide a novel opportunity to study the way natural fault systems respond to time-varying changes to the stress field. The recent 2024-2025 Fentale-Dofen dyking episode in the northern Main Ethiopian Rift (NMER) offers a rare opportunity to investigate these processes, as the surface deformation was captured in unusually high spatial and temporal resolution by satellite radar.
In our study, we combine Interferometric Synthetic Aperture Radar (InSAR) data from the COSMO-SkyMed satellite, high resolution Digital Elevation Model (DEM), with a catalogue of >150 relocated moderated-sized earthquakes (M4.5-6) to study the spatio-temporal evolution of seismic and aseismic fault slip linked to dyking in the NMER. We focus on an area ~15 km north of the tip of the dyke, where we find fault patches showing both repeated seismic and aseismic slip occurring in close proximity, associated with surface deformation of <13 cm in 3 months. We consider three possible mechanisms for the observed fault behaviour: (1) that this is normal mainshock-aftershock sequence on faults governed by rate-and-state friction, (2) that elastic stress perturbations from the ongoing dyke intrusions reloaded the fault patches, or (3) that elevated pore-fluid pressure caused transient reductions in effective normal stress on the faults. Using slip and stress modelling, we will test these hypotheses and quantify how much seismic/aseismic strain is accommodated by pre-existing and newly formed faults, as well as the relative contribution of seismic/aseismic strain to accommodating shallow crustal extension during the dyking episode.
These findings provide new constraints on fault mechanics and the interaction with magmatic processes in rifting environments, improving our understanding of dyke-induced seismicity and the evolving nature of faults with repeated earthquakes.
How to cite: Orrego, S., Biggs, J., Wimpenny, S., Zheng, W., Way, L., Vallée, M., Grandin, R., and Lewi, E.: Investigating seismic and aseismic fault motion caused by the 2024-2025 Fentale dyke intrusions in Ethiopia, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-12718, https://doi.org/10.5194/egusphere-egu26-12718, 2026.
Shallow slow slip events have long been observed along the strike slip faults of the San Andreas fault system and are now increasingly observed on many other faults on Earth. Creep events are thought to episodically release a portion of the fault’s interseismic stress budget that has accumulated over the earthquake cycle. However, it is not known what portion of strain these events release, and what residual strain remains available to drive earthquake occurrence. Near-field surface creep measurements, like alignment arrays and creepmeters, are unable to constrain the depth of creep, meaning that is difficult to constrain the release of strain or stress in each creep event without additional assumptions. In this study, we use radar data from InSAR platforms to resolve the depth of creep during creep events along the Superstition Hills fault in Southern California. We mitigate atmospheric noise by stacking co-event interferograms and by using empirically derived covariance matrices in the modeling. We apply a new nonlinear dislocation modeling method that constrains the slip distribution to be elliptical at each point along the fault and uses field and creepmeter data as lower bounds on surface slip. Using this model, we compute the strain drop throughout the rupture. We apply this technique to the 2006, 2010, 2017, and 2023 aseismic ruptures in Envisat, UAVSAR, and Sentinel-1 data. Lastly, we compare the resulting strain drops to strain accumulation rates calculated from backslip, testing the hypothesis that shallowly released strain is equal to the strain applied from deep dislocations in the crust. Using only the creep events in the instrumental record, we find that interseismic slip rates on the SHF must be above 10 mm/yr to explain the observations, a result consistent with regional-scale block modeling. Our results have implications for the strength of faults, the expected modes of seismic moment release in the shallow crust, and for seismic hazard analyses near creeping faults.
How to cite: Materna, K. and Bilham, R.: Shallow Aseismic Slip and Stress/Strain Budgets on the Creeping Faults in the Imperial Valley, California, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-21901, https://doi.org/10.5194/egusphere-egu26-21901, 2026.
Aseismic slip is increasingly recognized as a fundamental driver of earthquake nucleation, affecting the spatio-temporal evolution of seismicity, yet its direct observation remains rare due to limited strain measurements close to a natural fault system. Here, we present results from the 'FEAR1' experiment conducted at the Bedretto Underground Laboratory for Geosciences and Geoenergies (Switzerland), where we used fluid injections to activate a natural fault and fracture network in crystalline rock under in-situ stress conditions at ~1 km depth. This experimental setting is particularly well suited to investigate induced seismicity and the role of aseismic processes in fault activation, thanks to dense near- and on-fault strain, pressure, and seismic monitoring.
During several injections performed in FEAR1, we observed the activation of a steeply dipping, highly permeable fracture zone, which intersects a densely instrumented borehole. Hydraulic stimulations triggered seismicity (−4.9 < Mw < −2.3) that organized along a plane whose orientation is consistent with geological observations in boreholes cores, logs and on the laboratory tunnel wall. Simultaneously, high-resolution Fiber Bragg Grating strain measurements revealed progressive, irreversible tensile deformation localized near the fracture intersection with the monitoring borehole, reaching nearly 1000 µε over the course of the experiment.
Static elastic modeling demonstrates that the cumulative strain produced by the recorded earthquakes accounts for less than 1% of the observed deformation, indicating that fault slip was dominantly aseismic. The spatial and temporal evolution of seismicity shows systematic up-dip migration toward the strain concentration zone and the emergence of families of repeating earthquakes. The recurrence rate and cumulative slip of these repeaters correlate with the measured strain rate and strain, suggesting a scenario where seismic asperities are embedded within a creeping fault segment sustained by pore pressure stress perturbations.
Inversions of irreversible strain for simplified slip sources indicate a predominantly strike-slip mechanism consistent with the estimated local stress field, although trade-offs between source location, source dimension and slip direction highlight the limits of 1D strain observations. Our results provide direct experimental evidence for fluid-driven aseismic slip on a natural fault and demonstrate how microseismicity and repeaters can serve as indirect proxies for underlying slow deformation.
How to cite: Lambiase, A., Meier, M.-A., Spagnuolo, E., Nikkhoo, M., Marsan, D., Rinaldi, A. P., Gischig, V., Selvadurai, P., Cocco, M., Giardini, D., and Wiemer, S.: Fluid-induced aseismic slip and seismicity on a natural fault: insights from the FEAR1 experiment at BedrettoLab, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-14725, https://doi.org/10.5194/egusphere-egu26-14725, 2026.
Slow-slip events (SSEs) are a well-known mode of aseismic deformation in subduction zones. Seismological and geological studies further suggest that SSEs enhance fault-zone permeability, enabling fluid migration from overpressured oceanic crust into the plate interface. However, it remains unclear whether the resulting pore-pressure changes dominate stress transfer and promote the commonly observed increase in seismicity during SSEs and, although less commonly, the occurrence of larger megathrust earthquakes. Here, we investigate the impact of an SSE that occurred three days before the 2017 Mw 6.9 Valparaíso earthquake in central Chile. We use a forward 4D hydromechanical (poroelastic) model and compare the resulting spatial stress changes with a high-resolution seismicity catalog of the foreshock sequence.
We simulate the SSE by prescribing a geodetically inferred slip distribution on the fault interface and assume an overpressured oceanic crust, together with transient permeability enhancement due to SSE-induced local fracturing of the plate interface. We compute stress transfer driven by these pore-pressure changes along the plate interface and compare the results with widely used models that consider elastic stress changes only. Our results show that fluid migration into the plate-interface zone generates stress changes of ~1–10 MPa, overwhelmingly dominated by pore-pressure variations. The largest stress (and pore-pressure) changes spatially correlate with zones of increased seismicity, repeating earthquakes, and the mainshock. In contrast, the elastic-only scenario produces stress changes that are two to three orders of magnitude smaller and shows a much weaker spatial correspondence with the observed seismicity.
Our modeling results indicate that transient permeability enhancement during SSEs enables fluid redistribution that fundamentally controls stress transfer along the plate interface. We conclude that pore-pressure changes exert first-order control on earthquake precursors in subduction zones, offering a physical explanation for foreshock clustering and the triggering of large earthquakes during SSEs. These findings highlight the importance of incorporating fluid–rock interactions in models of seismic hazard and earthquake nucleation.
How to cite: Peña, C., Cabrera, L., Muñoz-Montecinos, J., Ruiz, S., and Heidbach, O.: Hydraulic control of the foreshocks and mainshock of the 2017 Valparaiso earthquake in central Chile, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-6156, https://doi.org/10.5194/egusphere-egu26-6156, 2026.
In subduction zones, the depth-dependent release of fluids from compaction and metamorphic dehydration reactions in hydrous lithologies plays a key role in modulating pore fluid pressure, fault strength, and slip behavior along the megathrust. The depth-distribution of fluid release is also the primary control on volatile fluxes through the forearc, and on the residual volatile content of the subducting plate. Here, we investigate the inventory and release of fluids from altered oceanic crust by low-grade dehydration reactions (~50-350 °C) at the Northern Hikurangi subduction zone, where slip on the outer (shallow) megathrust is accommodated almost entirely in frequent, large shallow slow slip events (SSEs).
Regional geophysical surveys and drilling during International Ocean Discovery Program (IODP) Expedition 375 show that the incoming plate of the Hikurangi Plateau carries a thick (>1.5 km) and extensively altered volcaniclastic sediment blanket characterized by an abundance of phyllosilicates (primarily Mg-smectite) and zeolite, and mineral-bound water contents as high as 14-16 wt.%, into the SSE source region. We quantify the distribution of fluid release from this sediment package by combining compaction trends to assess compactive water loss and thermodynamic phase equilibria models using sediment drill-core compositions to compute water release from dehydration reactions.
We find that: (1) compactive dewatering dominates in the outermost 15-20 km of the forearc, where temperatures remain too low (<100 °C) to drive dehydration reactions; and (2) a large volume (~5-8 wt.%) of mineral-bound water is released step-wise over the region spanning from ~30-90 km from the trench (corresponding to depths of 5-15 km below seafloor and temperatures of 150-260 °C), primarily from decomposition of zeolite and phllyosilicate phases. This contrasts with the behavior of Ca- and Na-smectites typically found in detrital marine sediments and altered volcanic ash, which undergo dehydration between 80-150 °C.
Because the majority of compactive dewatering precedes dehydration, mineral-bound water is released where porosity, permeability, and compressibility are reduced, maximizing the potential for excess pore pressure generation along and beneath the megathrust. The broad region of low-temperature metamorphic fluid release directly overlaps the slip zone of recurring SSEs, supporting the idea that dehydration - and associated elevated pore pressures and low effective normal stress - favor SSE as the prevailing mode of strain release on the plate interface. The presence of thick extensively hydrated oceanic crust and persistence of fluid production from clay dehydration to ~260 °C contrasts with other subduction zones, where low-T metamorphism is dominated by the transformation of Ca- and Na-smectites to illite by 120-150 °C. We speculate that this difference may offer an underlying explanation for the lack of a locked seismogenic zone at the Northern Hikurangi margin, whereas at other subduction margins, a lack of significant fluid production from dehydration in the 150-350 °C window may lead to a better-drained megathrust and promote stick-slip behavior.
How to cite: Saffer, D. and Smye, A.: Links between Low-T Dehydration and Recurring Shallow Slow Slip Events in the Northern Hikurangi Subduction Zone, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15863, https://doi.org/10.5194/egusphere-egu26-15863, 2026.
Fluid pressure variations within fault zones impact fault strength and have the potential to produce detectable geophysical signals that can help characterise fault dynamics. One key process impacting fluid pressure is pore volume variations (dilation or compaction) due to stress changes and inelastic deformation. Slip-induced dilation and compaction have been thoroughly documented in laboratory experiments, but their impact on pore pressure has not. In nature, we expect slip to be associated with stress variations, and there might be cumulated effects of poroelastic and inelastic pore pressure changes. In order to document such effects, we conducted laboratory rock friction experiments where fluid pressure was monitored in situ during sequences of quasi-static loading followed by dynamic slip event. The simulated fault was a 30 degrees saw-cut in a Westerly granite cylinder, saturated with water, tested under triaxial conditions. The low hydraulic diffusivity of the rock made the fault and wall rock transiently undrained during deformation. During quasi-static loading with no fault slip, we observed pore pressure rises that we interpret as poroelastic closure of the fault. During dynamic slip events, pore pressure systematically dropped, approximately in proportion to the drop in normal stress. A large contribution to the pore pressure drop is interpreted as poroelastic opening of the fault. Prior to stick-slip events, we detected systematic pore pressure decreases by up to around 1 MPa, correlated to the occurrence of inhomogeneous slip along the fault. Slip nucleation, inferred by kinematic inversion of local strain gauge data, is linked to local slip magnitudes of the order of 1 to 10 µm, and appears to lead to inelastic dilation. A stability analysis of fault slip including dilatant and poroelastic effects shows that poroelastic coupling tends to compensate normal stress variations, leading to faults operating under mostly constant effective normal stress if conditions are undrained.
How to cite: Brantut, N., Passelègue, F., and Dublanchet, P.: Pore pressure change during nucleation and slip along experimental faults, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-17727, https://doi.org/10.5194/egusphere-egu26-17727, 2026.
Slip instabilities leading to earthquakes, landslides, and glacier surges may be triggered by high fluid pressures. On the other hand, high fluid pressures also suppress instability because of large nucleation length-scales in overpressured systems. We review geological, glaciological, and rock mechanical observations and highlight two key scales that control pore pressure induced frictional instabilities: (1) the length scale over which pore fluid overpressure is maintained, and (2) a time scale defined by relative rates of deformation propagation and pore fluid transport. These scales are also dependent on rheological regime, and we find three end-member regimes: (1) shallow and/or low temperature deformation where ambient stress is low, faults are close to frictional failure, and shear is easily delocalised; (2) deformation in a brittle and frictionally unstable regime (such as the crustal seismogenic zone), where planes are close to frictional failure and slip tends to localise; and (3) environments where viscous deformation is preferred over frictional, and hence bulk stress is low, frictional strength is high, and delocalisation dominant. In regimes 1 and 3, fluid-driven instabilities tend to be confined to local areas of overpressure, because deformation delocalises in the bulk and dilatant hardening prevents further propagation. In regime 2, however, slip tends to localise and it is potentially favourable for fluid-induced instabilities to grow, provided slip surfaces are sufficiently close to failure. These regimes also apply to glaciers, where viscous flow of ice competes with frictional sliding on the glacier base - here, interconnected overpressured water at the glacier base is a commonly invoked mechanism that promotes frictional instability. These concepts imply that fluid-driven frictional instabilities are only as large as the areas where fluid overpressured patches can be interconnected, and therefore highlight the key role of fluid pressure heterogeneity in determining whether fluid-induced instabilities can propagate.
How to cite: Fagereng, A., Zhu, W., Gagliardini, O., V. Schuler, T., and Renard, F.: Pore Fluid Pressure Effects on Friction and Fracture, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-14991, https://doi.org/10.5194/egusphere-egu26-14991, 2026.
Rock deformation experiments play a key role in our understanding of earthquake physics and friction constitutive laws. These laws commonly describe the response of analogue laboratory faults as a simple and homogeneous system, without accounting for the spatial-temporal evolution of structures in the sample. However, increasing experimental evidence suggests that slip instability is closely tied to heterogeneity, complex rheologies, and inhomogeneous boundary conditions. To address this, we designed a novel transparent setup to observe real-time deformation, track the spatial-temporal evolution of shear fabric, and document unstable slip in experimental faults. Our video documentation reveals that the progressive development of fault fabrics results in heterogeneous but not random stress redistribution. We show that stress and structural heterogeneities play a key role in the nucleation, propagation, and arrest of slip instabilities, raising questions about the robustness of scaling laboratory frictional laws to nature.
How to cite: Pozzi, G., Volpe, G., Taddeucci, J., Cocco, M., Marone, C., and Collettini, C.: Spontaneous complexity in the dynamics of slow laboratory earthquakes., EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-3472, https://doi.org/10.5194/egusphere-egu26-3472, 2026.
After an initial phase of circular expansion, very large earthquakes primarily grow horizontally, with their vertical extent limited by the seismogenic width of the Earth’s crust. This geometric evolution is accompanied by a transition in rupture dynamics from crack-like to pulse-like propagation. Such events are commonly referred to as elongated ruptures.
While classical models (f.e., Linear Elastic Fracture Mechanics (Freund, 1998)) successfully describe small to moderate earthquakes, they fail to capture the dynamics of large events. Recent theoretical and numerical work by Weng and Ampuero (2019) introduced a physical framework for elongated ruptures, which, although supported by numerical validation and natural observations, has yet to be experimentally validated.
To address this, we conducted 2D rupture experiments in a biaxial direct shear apparatus under unbounded and bounded conditions. The unbounded case corresponds to a uniform velocity-weakening interface, while the bounded case consists of an elongated velocity-weakening region adjacent to a wide velocity-strengthening zone, mimicking a seismogenic layer whose width is bounded by deep aseismic regions. This experimental model successfully reproduces confined elongated ruptures and reveals distinct propagation styles: crack-like ruptures under unbounded conditions and pulse-like ruptures under bounded conditions. This transition is also reflected in the temporal evolution of seismic moment: during the initial phase of propagation, seismic moment scales cubically with rupture duration, while after saturation of the seismogenic width, it transitions to a linear scaling, as expected for pulse-like ruptures.
Together, these observations highlight the role of the seismogenic layer in controlling rupture style and provide experimental support for the proposed theory of elongated ruptures.
How to cite: Paglialunga, F., Ampuero, J. P., and Passèlegue, F.: Seismogenic width control on the dynamics and scaling of laboratory elongated ruptures, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-16488, https://doi.org/10.5194/egusphere-egu26-16488, 2026.
As the key mechanism of shallow earthquakes, the fault stick-slip behavior is usually explored under the assumption of constant normal stress. However, dynamic natural processes (tides, far-field earthquakes, etc.) and human activities (blasting, water injection, mining, etc.) generate periodic stress disturbances in the fault zone. So far the coupled results of fault seismic slips under variable normal stress are poorly understood.
We performed laboratory direct shear tests on saw-cut granite joints under constant and cyclic normal stress (σn), considering the role of load point velocity (Vlp), normal stress oscillation amplitude (ε) and normal stress oscillation frequency (f). Under constant normal stress, the joint exhibits a spontaneous stick-slip phenomenon for different Vlp. The shear stress drops and recurrence timespans of stick-slip events are reduced with faster Vlp. Under equivalent σn level, the cyclic σn weakens the frictional strength when Vlp is small and enhances the strength when Vlp is large. As ε grows, the joint slip style switches from regular stick-slip to chaotic slip, and eventually to compound stick-slip. The frictional strength is first increased and later weakened. In respect to effect of f: when f is small, one σn cycle can produce several stick-slip events. When f is medium, the period of the stick-slip event is equal to the cyclic σn period. For further increase of f , the recurrence period of stick-slip events becomes double the cyclic σn period. The frictional strength is decreasing or increasing at the critical point for frictional resonance.
The improved spring-block model equipped with rate-and-state friction framework matches the lab observations satisfactorily. Especially, the introduction of a stiffness response coefficient (Ψ) allows the model to reflect realistic fault frictional behavior, where shear stiffness varies with σn. A new parameter Θ is defined whose symbol (+ or -) directly determines the compression/relaxation status of the spring, and satisfactorily explains transitions in shear stress trends. Comparative analysis with the conventional Linker-Dieterich model highlights the improved physical consistency of our new approach, particularly in preserving the physical interpretation of the state variable, θ. The model also demonstrates that under large σn disturbances, a frictional system can effectively exhibit stick-slip behavior even in the velocity-strengthening scope. More importantly, the modeling implies that fast slip events greatly reduce the contact density of the fault interface. The contact state of the stick-slip joint/fault cannot be judged solely by σn. The contact area during the shear process is determined by both, the real-time σn level and the state variable θ.
How to cite: Tao, K. and Konietzky, H.: Experimental and modeling insights into fault stick-slip behavior under dynamic normal stress condition, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-2318, https://doi.org/10.5194/egusphere-egu26-2318, 2026.
Independent lines of seismic evidence suggest that pore fluid pressure at the depth range of episodic slow slip events (SSEs) may undergo periodic fluctuations synced with the SSE slip cycles. Here we develop a numerical simulation framework that integrates the SSE model governed by the rate- and state-dependent friction with Bayesian data assimilation to optimize time-variable fault friction parameters, using constraints from the northern Cascadia GNSS time series. We first conduct synthetic experiments to calculate surface displacement time series from the rate-state SSE model generated fault slip history with imposed Gaussian noise. Both frictional parameters, effective normal stress (normal stress minus pore pressure) and characteristic slip distance, converge to their true values in 5-10 iterations from the initial guesses that are 10-20% off from the true values, demonstrating the feasibility of the data assimilation framework. We then apply this framework to 2009-2020 GNSS time series that encompasses SSE cycles recorded at ~ 30 stations along the northern Cascadia subduction zone. We use a GNSS time series of 1000 days (~3 SSE cycles) in each inversion run to fully resolve the temporal changes in stress or friction; longer time series will cause inversion convergence issues due to the system nonlinearity. Within an inversion run, we choose a sliding time window of 9 months for each optimization epoch, which is a trade-off that on one hand includes sufficient information for the prediction of fault slip in the next time step and on the other hand allows temporal distinctions between the inter- versus intra-SSE time periods. Our inversion results show clear cyclic fluctuations in the optimized characteristic distance and effective normal stress values during SSE cycles. Specifically, effective normal stress increases (pore pressure drops) during the intra-SSE period; effective normal stress decreases (pore pressure increases) during the inter-SSE period. The pore pressure oscillation pattern is independent of whether the characteristic slip distance is time-invariant during data assimilation, but the converse does not hold. Our results are thus consistent with the proposed pore pressure build-up and release processes, i.e., fault-valve model, at the SSE depth range.
How to cite: Liu, Y. and Zhang, W.: Fault-valve behavior during slow slip cycles constrained using Bayesian data assimilation for a Cascadia subduction fault slip model, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-13013, https://doi.org/10.5194/egusphere-egu26-13013, 2026.
Pull-apart basins introduce extensional bends and strong along-strike heterogeneity into otherwise strike-slip systems, potentially altering slow slip, earthquake nucleation, and multi-segment rupture. The Princes Islands segment of the Main Marmara Fault (MMF) hosts a ~30-40 km pull-apart basin within the Marmara seismic gap south of Istanbul, where geodetic and geological observations suggest partial unlocking and complex rupture behavior. Yet the mechanical role of this extensional geometry in controlling fault slip and rupture remains poorly constrained. We perform three-dimensional quasi-dynamic simulations using PyQuake3D (Tang et al., 2025) to quantify the impact of the Princes Islands pull-apart basin on interseismic loading, slow-slip transients, dynamic rupture propagation, and earthquake recurrence along the MMF. We model a continuous fault surface with variable dip, bends, and segmentation, and prescribe depth-dependent effective normal stress, spatial frictional contrasts, and stress heterogeneity representative of extensional basin environments, guided by published geophysical constraints. Our results show that the basin exerts a first-order control on slip style and rupture outcomes through the competition between geometric unclamping, frictional heterogeneity, and stress structure. Extensional bends favor localized unlocking and recurrent aseismic or slow-slip episodes, which in turn modulate where dynamic ruptures nucleate. At the event scale, the basin can behave either as a rupture barrier or a rupture accelerator: in many realizations, ruptures do not continue smoothly across the basin but instead produce triggered seismicity via static stress transfer and re-nucleation near segment boundaries. Stress concentrations at geometric transitions primarily govern nucleation locations, while frictional contrasts regulate rupture persistence and arrest. These findings highlight that explicitly representing extensional fault structures is critical for assessing multi-segment rupture potential and time-dependent seismic hazard on the MMF near Istanbul.
Tang, R., Gan, L., Li, F., & Dal Zilio, L. (2025). PyQuake3D: A Python tool for 3‐D earthquake sequence simulations of seismic and aseismic slip. Journal of Geophysical Research: Machine Learning and Computation, 2(4), e2025JH000871.
How to cite: Osei-Tutu, D., Sopaci, E., and Dal Zilio, L.: How an extensional pull-apart basin modulates fault slip and earthquake rupture on the Main Marmara Fault, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-3420, https://doi.org/10.5194/egusphere-egu26-3420, 2026.
Elucidating fault zone processes during long-term seismic cycles is critical for mitigating earthquake hazards in intraplate regions. We investigated the hydro-mechanical evolution of a strike-slip branch of the Yangsan Fault, SE Korea, which bounds Triassic and Jurassic granites. By integrating multiscale observation with high-velocity rotary shear experiments and XRD, we characterized the fault architecture, which consists of a <35 m thick damage zone surrounding a <1 m thick core. The core contains breccia and foliated gouge rich in clay minerals (43 wt.%), specifically dominated by illite (21.2 wt.%) and smectite (13.3 wt.%). Shear experiments on the foliated gouge revealed a consistently low friction coefficient (μss<0.17). Notably, instantaneous flash dilation of the mixed smectite/illite gouge was observed at seismic slip rates (1.3 m/s) when total displacement exceeded ~5 m. Microstructural cross-cutting relationships indicate a distinct sequence of events: (1) vigorous injection of pressurized fluids from wall rocks into the densely packed, low-permeability gouge directly; (2) precipitation of fibrous calcite veins along foliation planes and perpendicular to the Y-shear direction; and (3) subsequent injection of fluidized gouge material into the damaged wall rock. These observations suggest that cyclic coseismic and aseismic faulting occurred following the low-temperature formation of expanding clay minerals. We conclude that the dynamic interplay between fluid pressurization and the fluidization properties of clay gouge acts as a primary driver of mechanical instability, playing a key role in the long-term seismic evolution of intraplate granitic fault zones.
How to cite: Kim, C.-M., Woo, S., Lee, J., and Choi, J.: Fluid-Driven Injection and Pressurization of Clay-Rich Gouge in the Yangsan Fault: Implications for the Long-Term Seismic Cycle, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-4626, https://doi.org/10.5194/egusphere-egu26-4626, 2026.
Fluid-induced fault reactivation and associated seismicity is a critical process in reservoir exploitation and emerging geo-energy activities such as Carbon Capture and Storage (CCS), Enhanced Geothermal Systems (EGS) and wastewater disposal. During fluid injection, the fault stress state progressively approaches the failure criterion τ = (σₙ - Pf) * µ + C , where τ is shear stress, σₙ normal stress, Pf fluid pressure, µ friction, and C cohesion. Once the stress state reaches the failure envelope, faults may reactivate either seismically or aseismically. However, the mechanisms governing aseismic versus seismic fault reactivation during fluid injection remain debated.
Previous laboratory studies suggest that this seismic vs. aseismic deformation may be influenced by fault frictional properties influenced by mineralogy, fault zone structure, stress state, and injection rate, yet the relative contribution of these factors remains unclear. To address this issue, we present an experimental study on binary and ternary fault gouges with variable fractions of quartz, calcite, and illite. These are minerals found along faults zones and within reservoir rocks commonly exploited for geo-energy applications.
For each mineralogical composition, two experimental datasets were acquired. In the first dataset, we performed slide–hold–slide and velocity-step tests to measure friction, frictional healing and the velocity dependence of friction. In the second dataset, we investigated fault slip behavior during fluid pressure-induced reactivation at three different stress states.
The frictional properties reveal a pronounced contrast between granular and platy phyllosilicate-rich gouges. Granular materials exhibit high friction (µ ≈ 0.6), positive frictional healing, and low a–b values, indicating velocity-weakening and potentially seismogenic behavior. In contrast, illite-rich gouges (illite > 40%) display low friction (0.28 < µ < 0.4), low to negative healing, and strongly positive a–b values, indicative of velocity-strengthening and aseismic behavior. Duringfluid injection induced-reactivation, granular-rich gouges reactivate through an exponential increase in slip velocity, mimicking seismic-like instability. Conversely, illite-rich gouges reactivate through aseismic but accelerated creep that does not evolve into dynamic failure.
Notably, reactivation in granular gouges is abrupt and occurs at stress states well above the predicted failure envelope, whereas in illite-rich gouges reactivation is gradual and occurs at or before the predicted failure envelope. In addition, at constant illite content, quartz-rich gouges reactivate faster than calcite-rich fault gouges.
The integration of these results suggests a conceptual framework in which fluid-induced fault reactivation is governed by the interplay between frictional healing and rate dependence, with mineralogy exerting a first-order control. In granular gouges, strong healing dominates the the fluid induced reactivation process, leading to delayed but abrupt fault reactivation that can overcome the stabilizing slight rate-strengthening effect, promoting an exponential acceleration under fluid pressurization. In contrast, in phyllosilicate-rich gouges, weak or negative healing combined with a marked rate strengthening behavior stabilizes slip, favoring continuous aseismic creep.
This framework demonstrates that the balance between healing and rate dependence, strongly linked to fault mineralogy, governs whether fluid-induced fault reactivation produces seismic slip or aseismic creep.
How to cite: Coppola, L., Volpe, G., Giorgetti, C., Pozzi, G., Wibberley, C., Bourgeois, F., and Collettini, C.: Fluid induced fault slip behavior: frictional healing vs velocity dependence of friction, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-9583, https://doi.org/10.5194/egusphere-egu26-9583, 2026.
To address the limitations of traditional convolutional neural networks (CNNs) in seismic fault identification—such as restricted local receptive fields, limited capability for modeling long-range structural correlations, and low sensitivity to small or subtle faults—this study proposes a seismic fault identification framework based on a Vision Transformer (ViT) architecture combined with self-supervised pretraining and transfer learning. Self-supervised pretraining is first conducted on large volumes of unlabeled three-dimensional seismic data to learn general representations of geological structures, thereby reducing the dependence on manually labeled samples. The pretrained ViT model is subsequently transferred to the fault identification task and systematically compared with a conventional U-Net architecture. Experiments on a publicly available synthetic seismic dataset show that the ViT-based model achieves improved fault localization accuracy, spatial continuity, and robustness to noise compared to U-Net. Application to real 3D seismic data from an oilfield further demonstrates that the proposed method is capable of detecting a larger number of small-scale faults with enhanced structural continuity, highlighting its applicability in structurally complex settings. The results suggest that Transformer-based global modeling provides an effective alternative for automated seismic fault interpretation.
How to cite: Pei, H. and Zhang, G.: Seismic Fault Identification Based on Vision Transformer, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-4788, https://doi.org/10.5194/egusphere-egu26-4788, 2026.
At the northern Hikurangi margin, Aotearoa New Zealand, slow slip events (SSEs) recur every 6-24 months to ~30 km depth. While shallow SSEs (0-10 km) are well-studied offshore, the deeper portion (10-30 km) remains poorly understood, limiting insight into SSE initiation. In Woodward et al. 2026 we investigate the relationships between newly resolved SSEs and seismicity. We combine passive seismological, geodetic, geochemical and seismic reflection data to analyse the relationships between seismicity and slow slip events, and the mechanisms that invoke them. Using time-dependent inversions, we resolve two small SSEs (MW 6.2 and 6.4), one of which extends unusually deeply from 15 to 30 km depth. Using data from a dense onshore seismograph network, deployed directly above this deeper portion from December 2017 to October 2018, we construct a catalog of 3,071 high-quality earthquakes with hypocentral uncertainties ≤5 km, located with a 3-D velocity model and our new 1-D model. Focal mechanisms reveal numerous normal-faulting earthquakes, including some within the slab mantle. Seismicity distributions and normal-faulting earthquakes occur along vertically aligned pathways that link the subducting slab mantle to surface seeps, where fluids show mantle-derived signatures. We infer that normal faults form due to slab bending and localized uplift of subducting seamounts, which enhance plate interface roughness, damage the upper plate, and promote fluid migration. Landward of ~100 km from the trench, both surface seeps and normal-faulting mechanisms cease, coinciding with the downdip limit of shallow SSEs. Together, these results suggest that the Hikurangi margin’s rough subducting plate interface exerts strong control on forearc dewatering and SSE genesis.
How to cite: Woodward, A., Bastow, I., Bell, R., Wallace, L., Jacobs, K., Henrys, S., Fry, B., Merry, T., Lane, V., Ville, L., Houldsworth-Bianek, P., and Broadley, L.: The role of seamounts, fluids, and normal faults in slow slip regions: seismological insights from the northern Hikurangi margin, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-9842, https://doi.org/10.5194/egusphere-egu26-9842, 2026.
Aseismic slip plays a key role in earthquake dynamics, but we currently do not fully understand why some faults slide aseismically. Aseismic slip is largely influenced by the fault zone's frictional behaviour, by its material composition, the presence of fluids, the geometry of the fault, and the fault zone fabric. Recent research has focused on the material composition, and more specifically on the evolution of resistance to slip with slip speed for different types of rocks. Generally, faults with rate-weakening behavior tend to host earthquakes, while faults with rate-strengthening behavior accommodate stress through aseismic slip. However, even in rate-weakening materials, low effective normal stress, induced by high pore fluid pressure, makes it unlikely for a slip instability to reach the critical size needed to nucleate regular earthquakes. Hence, the presence of high pressure fluid within fault zones may explain the presence of shallow aseismic slip along faults. However, we lack direct evidence of the presence of fluids along various faults where aseismic slip has been identified.
We use data from a dense network of seismometers along the North Anatolian Fault Zone SEISMENET1 to investigate spatial changes in seismic velocity along the section hosting aseismic slip. This section slips aseismically since at least 1944 and is the epicentral region of the last two large earthquakes that have struck the area, namely the 1944 M7.3 Bolu-Gerede and the 1943 M7.6 Tosia-Ladik earthquakes. Using local earthquake tomography, we test for a possible presence of fluids in the fault zone and a damage zone surrounding the epicentral region of the 1943 and 1944 earthquakes.
Our network includes 5 broadband seismometers and 10 geophones deployed around the creeping section along a narrow swath paralleling the fault trace. In addition to data from these 15 temporary stations, seismic data from 5 permanent broadband stations were collected. The final dataset includes 24,756 P arrivals and 21,311 S arrivals from 2,272 earthquakes, with magnitudes ranging from Mw0 to Mw4. We use the simul2023 code2 to simultaneously determine the 3D structure of the shallow crust and relocate the earthquake hypocenters.
We find kilometer-scale shallow high vp/vs anomalies (values in range of 1.8 up to 1.95), consistent with a damage asymmetry aligned with the observed rupture directions of historical earthquakes, indicating long-term preferred rupture directions along this segment of the NAF. Additionally, the creeping section of the North Anatolian Fault is shown to correspond to a 30-km-long zone of vp/vs above 1.8, consistent with the presence of high pore fluid pressure within the fault zone. The findings provide compelling evidence that fluid processes, rather than fault zone rheology alone, significantly influence aseismic slip behavior along the NAF. Together, these results suggest a dynamic interplay between structural damage, rupture history, and fluid migration in controlling fault zone mechanics, with implications for improving seismic hazard assessment in creeping fault segments.
1https://geofon.gfz.de/waveform/archive/network.php?ncode=1O&year=2022
2https://doi.org/10.5281/zenodo.10695070
How to cite: Szrek, J., Jolivet, R., Schurr, B., Becker, D., Martínez-Garzón, P., Jara, J., and Çakir, Z.: Fluids and Creeping Faults: Insights From Local Earthquake Tomography of the Creeping Section of the North Anatolian Fault, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-12855, https://doi.org/10.5194/egusphere-egu26-12855, 2026.
Understanding the physical mechanisms governing megathrust seismicity and the geodynamic feedback between the megathrust and the overriding accretionary wedge remains critical in subduction zone geophysics. The structural complexity of accretionary wedges—characterized by heterogeneous porosity, permeability, and fault networks—critically influences the configuration of pore fluid pressure and frictional properties along the megathrust interface. To investigate these interactions, we employ a fully coupled hydro-mechanical numerical model (Gerya, 2019) that simulates two distinct timescales within a single, consistent rheological framework. Our approach incorporates temperature-dependent dehydration reactions, including smectite-to-illite and zeolite-to-greenschist transitions, to evaluate how fluid production and migration evolve during both subduction and seismic processes. Additionally, we implement a dynamic fault-valving mechanism where reference permeability evolves transiently to mimic fracture-induced permeability enhancement during fast slip. The simulation follows a two-stage workflow: first, we conduct long-term wedge accretion modeling with adaptive time steps (10–500 years) using a higher stress tolerance to construct realistic wedge architectures. Subsequently, we switch to a rupture simulation mode by reducing the stress tolerance, allowing the adaptive time-stepping scheme to automatically resolve short-term seismic cycles (from days to years). This methodology successfully introduces the structural and hydrological complexity inherited from long-term geological evolution into the analysis of short-term megathrust slip behaviors. Results indicate that fast slip events preferentially initiate at the transition zones between low and high overpressure regions, whereas domains characterized by high pore fluid pressure ratios () predominantly host slow slip events. Furthermore, we find that hydraulic properties control the spatiotemporal stability of rupture nucleation: higher permeability promotes significant temporal pore pressure variability, resulting in scattered initiation depths, while lower permeability maintains stable pressure configurations, leading to spatially consistent rupture nucleation. We conclude that the long-term hydro-mechanical evolution of the wedge governs megathrust nucleation and slip segmentation.
How to cite: Lin, C.-H. and Tan, E.: Hydro-Mechanical Modeling of Fluid-Regulated Deformation in Accretionary Wedges and Its Implications for Megathrust Slip, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-12927, https://doi.org/10.5194/egusphere-egu26-12927, 2026.
Given the significant risk that earthquakes pose to society, understanding the spatiotemporal evolution of slip rates on natural faults has been a central research objective over recent decades. Geological and geophysical observations indicate that fault slip is accommodated by multiple deformation mechanisms operating both on the fault plane and within the surrounding damage zone. At the outcrop scale, structures formed by seismic and aseismic slip commonly coexist within the same fault system, implying that fault displacement involves a combination of deformation processes controlled by mineral-specific rheology. At larger scales, however, these processes may be masked by the limited spatial and temporal resolution of geophysical and geodetic observations.
From a theoretical and experimental perspective, rate-and-state friction (RSF) laws have been widely used to explain unstable fault slip through velocity-weakening behaviour, in which fault strength decreases with increasing slip rate. In this framework, fault friction is governed by slip velocity and a single state parameter that evolves with time and slip, commonly expressed either through an aging law, where fault healing occurs primarily during stationary contact, or a slip law, where state evolution is driven by slip-dependent renewal of contacts. Because both formulations are typically expressed as local, slip-rate-dependent laws, the explicit role of cumulative slip history in controlling the fault state remains implicit.
To investigate the effect of cumulative slip history, we perform a suite of 3D quasi-dynamic simulations assuming a homogeneous distribution of rate-weakening frictional properties, in which fault slip is governed by a classical RSF formulation. By systematically decreasing effective normal stress from 50 to 10 MPa and explicitly rewriting the aging law in a slip-dependent, event-integrated form, we show that the well-documented transition from characteristic earthquake behaviour to deterministic chaotic slip (e.g., Rubin, 2008; Cattania, 2019; Barbot, 2019) is accompanied by a change in the role of the state variable. Specifically, state evolution becomes increasingly governed by cumulative slip history and slip-filtered healing, which inhibits convergence toward a unique healed state and results in incomplete state recovery between successive events. Importantly, this behaviour arises without prescribing any spatial heterogeneity in frictional properties or fault-zone structure.
These results have direct implications for the interpretation of fault kinematics. While regions where slip rates remain below the prescribed background velocity may persist constant over interseismic periods, the shear stress within those regions in models of low effective normal stress need not be stationary. Instead, shear stress can evolve significantly because of incomplete state recovery driven by cumulative slip history and slip-limited healing, leading to temporally heterogeneous mechanical behaviour despite stable kinematic expression.
References
Rubin, A. M. (2008). Episodic slow slip events and rate‐and‐state friction. Journal of Geophysical Research: Solid Earth, 113(B11).
Cattania, C. (2019). Complex earthquake sequences on simple faults. Geophysical Research Letters, 46(17-18), 10384-10393.
Barbot, S. (2019). Slow-slip, slow earthquakes, period-two cycles, full and partial ruptures, and deterministic chaos in a single asperity fault. Tectonophysics, 768, 228171.
How to cite: Julve Lillo, J., Fagereng, A., Ampuero, J.-P., van den Ende, M., and Toffol, G.: Fault state evolution governed by cumulative slip history., EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-14122, https://doi.org/10.5194/egusphere-egu26-14122, 2026.
Laboratory investigations into the behavior of fault zones have been a significant focus in experimental rock mechanics over the past decades. Various approaches have been developed, ranging from analog models to testing natural samples in triaxial cells. The primary goal of the latter is to infer the physical mechanisms responsible for failure under realistic conditions encountered in natural settings, albeit on small sample sizes (e.g., centimeter scale). In contrast, analog modeling aims to replicate similar mechanical behavior by applying scaling laws to geometry and material properties. In the present study we aim at combining large scale experiments (meter scale) with small scale experiments (cm scale) in order to highlight the underlying physical mechanisms preceding the slip.
To address the spatial scale limitations of classical rock mechanics, we developed new experiments that bridge the gap between traditional rock mechanics and analog experiments. These experiments utilize the unique capabilities of the DIMITRI setup, a giant true-triaxial apparatus (1.5m × 1.5m × 1m). Due to the size of this experimental device, the maximum stress it can apply is limited to 2 MPa per principal stress. Consequently, we chose polystyrene as an analog for rocks. The low elastic properties of polystyrene slow down physical processes, enabling comprehensive observation of rupture phenomena, from initiation to failure arrest. Our objective is to investigate the interplay between different types of slip occurring along the interface.
In this study, we conducted stick-slip experiments on large-scale polystyrene blocks with a pre-cut surface area of 1.5 m². We applied shortening rates ranging from 1 to 10 mm/min. Our experiments successfully reproduced stick-slip behavior, allowing us to observe variations in frictional behavior along the interface and identify different types of slip, from slow slip to dynamic slip.
In parallel, we performed small scale experiments uniaxial stick slip experiments, under the same conditions than the previous, that we monitored with 2D X-ray radiography at high frequency (12Hz). Preliminary observations highlight density contrasts in the bulk material around the fault plane, offering insight into potential precursory signs of slip.
Therefore, this study including two scales of observation demonstrates the relevance of our material to study the physical mechanisms controlling various slip types occuring along the seismic cycle.
How to cite: Bonnelye, A., Walter, B., Gouedar, A., and Faure-Catteloin, D.: PolystyQuakes : what can we learn from the use of polystyrene as analogue to earthquakes? From small to large scale and vice versa., EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-14191, https://doi.org/10.5194/egusphere-egu26-14191, 2026.
The Upper Tiber Valley, in the Northern Apennines of Italy, is a key natural laboratory for investigating how continental extension is partitioned between seismic and aseismic deformation. Extension in this sector of the Apennines has been primarily accommodated by the Altotiberina Fault (ATF), a low-angle (~15°) normal fault that is mechanically unfavorable for elastic shear failure, and by a network of high-angle synthetic and antithetic faults in its hanging wall. While the ATF is characterized by persistent background micro-seismicity, the high-angle faults host larger historical earthquakes and frequent seismic swarms, likely induced by fluid circulation and elevated pore pressure. Since the study of Anderlini et al. (2016), the local GNSS network has been significantly densified within the framework of the Alto Tiberina Near Fault Observatory (TABOO-NFO). The updated dataset now better resolves a sharp ~3 mm/yr chain-normal interseismic velocity gradient across the Upper Tiber Valley, providing unprecedented constraints on how ongoing extension is distributed across the fault system. We use the new GNSS velocity field to reassess the relative contribution of low-angle versus high-angle faults to crustal deformation and to quantify the partitioning between seismic and aseismic slip. We apply a block-modeling approach that jointly estimates rigid block rotations and spatially variable interseismic coupling through a newly developed iterative inversion strategy. The model includes 3D geometries, discretized in triangular dislocation elements, of both the ATF and its antithetic structures, permitting assessment of distributed slip rates across the fault system. Preliminary results show that shallow locking on high-angle syn- and antithetic faults plays a first-order role in explaining the observed velocity gradient, whereas the ATF accommodates a significant fraction of extension through aseismic creep. These findings refine earlier interpretations and provide new insight into how low-angle normal faults can interact with steeper faults during the earthquake cycle.
How to cite: Serpelloni, E., Nucci, R., Poggiali, G., Buttinelli, M., Anderlini, L., Marone, C., and Chiaraluce, L.: Seismic vs aseismic deformation in the Northern Apennines constrained from dense GNSS velocities, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-20370, https://doi.org/10.5194/egusphere-egu26-20370, 2026.
The Main Marmara Fault lies at the western termination of the North Anatolian Fault. While the North Anatolian Fault ruptured from its eastern termination to the Izmit segment over the 20th century through a westward‐propagating sequence of Mw ≥ 7 earthquakes, the underwater Marmara segment has not experienced a large earthquake in recent times. However, due to its proximity to the megacity of Istanbul, this segment represents one of the most hazardous fault systems in the Middle East. In particular, no historical earthquake has been identified on the Central Basin segment since at least 1766, potentially making it a major seismic gap.
To better assess seismic hazard along the Main Marmara Fault, we estimate the slip rate by jointly using a dense GNSS velocity field and four Sentinel-1 InSAR tracks. The GNSS velocity field consists of 111 measurements, including newly acquired densified sites along the northern shore of the Marmara Sea. The InSAR velocity field was processed automatically within the framework of the FLATSIM project, covering the period from October 2016 to April 2021. InSAR velocities are referenced to a Eurasia-fixed plate using the GNSS velocity field. We then perform a joint Bayesian inversion of slip rates using both datasets, allowing us to quantify uncertainties on the estimated slip rates.
Our results indicate that the Main Marmara Fault is predominantly creeping between longitudes 27.5 and 28.6, implying that the Central Basin segment is largely aseismic. Uncertainty estimates and forward modeling demonstrate that our datasets are capable of resolving slip behavior on this segment with good accuracy. However, the shallow portion of the Central Basin segment is still accumulating up to ~10 mm/yr of slip deficit, which could permit earthquakes of up to Mw 6.0 every few decades, similar to the 2025 sequence. West of longitude 27.5 and east of longitude 28.6, including the Prince Islands segment, the fault appears to be mostly locked down to 12 km depth. On the Prince Islands segment, close to Istanbul, the accumulated strain has the potential to generate an earthquake with Mw > 7.
How to cite: Klein, É., Neyrinck, E., Rousset, B., Masson, F., Ozkan, A., Yavasoglu, H. H., Ulrich, P., Jolivet, R., Doubre, C., Durand, P., and Doin, M.-P.: Slip Rates on the Main Marmara Fault from Bayesian Inversion of Dense GNSS and InSAR Velocity Fields, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-18025, https://doi.org/10.5194/egusphere-egu26-18025, 2026.
