Ares J. Rosakis

Abstract This article investigates the influence of the rotational components of ground motion on the seismic response of structures. Two independent seismic stations, located 30 m apart and within 1 km of the fault rupture, recorded the near-field motions of the 6 February 2023 Mw 7.8 Kahramanmaraş–Pazarcık earthquake. A third, synthetic station was generated using a stochastic frequency-domain approach to estimate the complete set of rotational components. The synthetic records assume that spatial variations in motion depend solely on the interstation separation distance. The resulting rotational response spectra indicate limited sensitivity to the random variability introduced by the synthetic station. For the first time, near-fault rotational ground-motion components—pitch, roll, and yaw—derived from a major earthquake are estimated and used as an input in nonlinear response-history analyses of 3D reinforced concrete building models. The case studies show that including rotational components increases peak core-wall shear forces by 5.3% and peak interstory drift by 13.3%. These findings underscore the importance of measuring and incorporating rotational ground-motion effects in seismic design codes and guidelines to better mitigate the vulnerability of structures near active faults.

Failure models related to dynamic slip pulses have been extensively used in seismology as for example the models of Yoffe (1952), Freund (1979) and Rice et al (2005). The models of Freund and Rice et al. describe a finite mode III crack of constant length that propagates at constant (subsonic) rapture velocity whereas the trailing edge leaves behind a dislocation as a permanent irreversible deformation. The driving forces of such models are far field shear tractions, whereas at the same time a frictional traction acts along the crack line due to the contacting crack surfaces that allow only shear slip. Yoffe’s model on the other hand assumes a complete healing and thus rendering the trailing edge to behave as a crack-tip, just as the leading edge does. An important difference between Freund’s and Rice et al.’s models is that Rice et al.’s model includes additionally the effect of the static friction, whereas Freund’s model incorporates only the kinetic friction. The present work extends the models of Freund and Rice et al. to flexoelectric materials under steady state anti-plane loading. The problem is formulated so that the mechanical problem is casted as a strain gradient elasticity problem. Our analytical results point out that the well known influence of the cohesive zone in classic dynamic fracture can be established in the context of flexoelectricity (and strain gradient elasticity) without the need for a cohesive zone. An interesting analogy with the Rice et al. analysis can be established and, in this way, a flexoelectric material can be used for laboratory work in earthquake and friction studies. • FEM analysis of dynamic slip pulses • Flexoelectric dynamic slip pulse • Asymptotic analysis for Flexoelectric dynamic slip pulse • Energy Release Rate for Flexoelectric slip pulse • Flexoelectric slip pulse analogue with cohesive zone model

Abstract Strike-slip faults—where tectonic plates grind past each other horizontally—are a defining feature of many densely populated continental seismic zones worldwide, including the San Andreas fault system in California, the North and East Anatolian faults in Türkiye, and the Sagaing fault in Myanmar (Burma). Although their lateral motion has long been recognized, a growing body of global evidence is now highlighting a more hazardous aspect of these systems: supershear earthquakes—fast propagating ruptures that exceed the speed of shear waves and can cause disproportionately intense shaking and destruction. Four of the last six Mw 7.0+ earthquakes on strike-slip faults have been recognized as supershear events, including the damaging Mw 7.7 Myanmar and the Mw 7.8 Pazarcik earthquakes, highlighting the need to confront the potential implications of such future events.

We have developed a coupled computational mechanics framework that integrates fully 3-D models for earthquake rupture dynamics with fluid mechanics models of tsunami generation and propagation. The three-dimensional, time- dependent, vertical and horizontal ground motions from spontaneous dynamic rupture models are translated into a moving bathymetry of the bay that drives the 2D nonlinear shallow water-wave equations. We find that supershear ruptures propagating along underwater, strike-slip faults, traversing narrow bays, are prime candidates for tsunami generation. We also show that, the dynamic focusing effect and the large horizontal displacements, characteristic of strike-slip earthquakes (especially super-shear ones) on long faults, are the critical drivers for the tsunami hazard in bays. These findings point to intrinsic mechanisms for tsunami generation by strike- slip faulting that do not require us to invoke complex seismic sources, landslides, or complicated bathymetries. We identify three distinct phases in the tsunami motion; an instantaneous dynamic phase, a lagging co-seismic phase, and a classical post-seismic phase, each of which affect the coastal areas differently. We conclude by emphasizing the need for re-evaluating the near-source tsunami hazard to coastal areas (e.g. the SF bay area in CA, the sea of Marmara near Istanbul, or the bay of Al Aqaba in the Red Sea) from local strike-slip faults.

Abstract Rupture speed is a crucial parameter of earthquake dynamics and influencing associated seismic hazards. Accurately resolving the rupture evolution of large earthquakes is essential for identifying factors governing earthquake physics. In this study, we investigate the kinematic rupture processes of the 2023 Mw7.8 and Mw7.7 eastern Turkey earthquake doublet. We integrate various complementary data sets and methods, including 3D surface deformation, teleseismic back‐projection, near‐fault strong motion waveform analysis, and finite fault inversions, to resolve the rupture details. Our results reveal that the Mw7.8 earthquake predominantly involves an asymmetric bilateral rupture on the main fault, with part of the northeastward rupture reaching super‐shear speed (∼5.2 km/s), while the southwestward rupture propagates primarily at the generalized Rayleigh speed (∼3.4 km/s), a characteristic of an inhomogeneous fault zone separating two dissimilar materials. This directional dependence on rupture speed may be attributed to a material contrast between the softer Anatolian plate and the stiffer Arabian plate, as supported by the fault zone head wave observations and tomography models. In contrast, the Mw7.7 event features a bilateral super‐shear rupture, likely due to its occurrence on intraplate faults without substantial material contrast across the fault. This study underscores the importance of incorporating detailed fault zone structures and high‐quality near‐fault observations into earthquake physics and seismic hazard analysis.

SUMMARY We investigate the impact of sediment layers on ground motion characteristics during subshear and supershear rupture growth. Our findings suggest that sediment layers may lead to local supershear propagation, affecting ground motion, especially in the fault parallel (FP) direction. In contrast to homogeneous material models, we find that in the presence of sediment layers, a larger fault normal (FN) compared to FP particle velocity jump, reflects shear propagation at depth but does not rule out shallow supershear propagation. Conversely, a large FP compared to FN particle velocity jump indicates supershear propagation at depth. In the presence of a shallow layer, we also uncover a non-monotonic behaviour in the sediment’s influence on supershear transition and ground motion characteristics. During supershear propagation at depth we observe that sediment layers contribute to enhancing FP velocity pulses while minimally affecting the FN component. Furthermore, in the limit of global supershear propagation we identify local supersonic propagation within the sediment layers that significantly alters the velocity field around the rupture tip as observed on the free surface, creating both dilatational and shear Mach cones. In all our models with sediments we also find a significant enhancement in the fault vertical component of ground velocity. This could have particular implications for hazard assessments, such as in applications related to linear infrastructure, or a higher propensity to tsunami wave generation. Our research unravels the importance of considering heterogeneous subsurface material distribution in our physical models as they can have drastic implications on earthquake source physics.

This work investigates the deformation and polarization fields around a finite anti-plane shear (mode III) crack growing dynamically under steady-state conditions. The leading tip of this finite crack breaks the material while the trailing tip heals it. This fast moving finite crack (referred to as a rupture “ pulse ” in the geophysics literature) propagates with a constant velocity and with the mechanical and the electrical fields that remain invariant with respect to an observer moving with the crack-tips. This problem belongs to the first type of steady state crack growth problems according to the classification of Freund. The “ prototype ” problem which refers to an isotropic, body subjected to fracture under tensile loading was first proposed and solved by Yoffe, while finite cracks (or shear pulses) were also analyzed by Freund and by Rice. In the above cases the material was assumed to be linear elastic. Our analysis extends these studies to flexoelectric materials, and it is both theoretical and numerical. It discusses the asymptotic structure of the crack-tip displacement and the polarization fields; it calculates the dynamic energy release rate and presents their dependence on crack-tip velocity. Comparisons are made to the available, classical, elasto-dynamic solutions and to the static case. The influence of the electrical properties of the material on strengthening is also analyzed. Dynamic fracture of flexoelectric materials is of relevance to both the study of earthquake source mechanics and to the analysis of the reliability of micro-electronic devices. This is because both rocks and ceramics are flexoelectric. Indeed, during earthquake rupture processes, dynamic, in-plane shear (Mode-II) and out of plane shear (Mode-III), cracks propagate along faults and exhibit both mechanical and electrical polarization signatures. At an entirely different length scale, flexoelectric ceramics are currently used as sensors and transducers and can experience dynamic shear failure along interfaces when subjected to dynamic loading (e.g. impact.). Failure by dynamic fracture can be detrimental to both their mechanical reliability and electrical functionality.

We investigate the impact of sediment layers on ground motion characteristics during sub-Rayleigh and supershear rupture growth. Our findings suggest that sediment layers may lead to local supershear propagation, affecting ground motion, especially in the fault parallel (FP) direction. In contrast to homogeneous material models, we find that in the presence of sediment layers, a larger fault normal (FN) compared to fault parallel (FP) particle velocity jump, reflects sub-Rayleigh propagation at depth but does not rule out shallow supershear propagation. On the other hand, a large fault parallel (FP) compared to fault normal (FN) particle velocity jump indicates supershear propagation at depth. In the presence of a shallow layer, we also uncover a non-monotonic behavior in the sediment’s influence on supershear transition and ground motion characteristics. During supershear propagation at depth we observe that sediment layers contribute to enhancing FP velocity pulses while minimally affecting the FN component. Furthermore, in the limit of global supershear propagation we identify local supersonic propagation within the sediment layers that significantly alters the velocity field around the rupture tip as observed on the free surface, creating both dilatational and shear Mach cones. In all our models with sediments we also find a significant enhancement in the fault vertical component of ground velocity. This could have particular implications for hazard assessments, such as enhanced linear infrastructure demands, or a higher propensity to tsunami wave generation. Our research unravels the importance of considering heterogeneous subsurface material distribution in our physical models as they can have drastic implications on earthquake source physics.

The dispersion relations in flexoelectricity are examined for plane time-harmonic waves that propagate in the flexoelectric materials. In contrast to classic elastodynamics, dispersion is observed in the displacement field due to two micro-structural and two micro-inertial lengths that emerge from the electromechanical coupling. In the absence of such coupling, we return to the classic elastodynamic results. The problem dissociates in longitudinal and transverse waves, as is the case in classic elastodynamics. The group velocity of the mechanical field is also the velocity of the energy transfer across the planes of the waves. An optical branch of the dispersion relation appears due to the polarization field that follows the mechanical field. The longitudinal and transverse velocities of the plane waves was found to depend on the corresponding microstructural lengths and are less than or equal to the classic plane wave velocities because the micro-inertial lengths are greater than or equal to the micro-structural length. The opposite effect is expected when we encounter flexoelectric metamaterials in which case the micro-inertial lengths are less than the micro-structural length.

This paper discusses the results of 1∶25 scale shake table tests evaluating the seismic response of multiblock tower structures (MTSs) conceived as energy storage systems. The tests described here are a part of a comprehensive research campaign involving smaller physical models, computational model validation, and the theoretical background required to compare results across scales. The 6.46-m-high MTSs consisted of over 7,000 concrete blocks stacked vertically without any bonding agent, interacting only by friction and rocking. Three MTSs were tested under two different ground motions. Dynamic digital image correlation (DIC) and low-cost micro electrical mechanical system (MEMS) accelerometers were used for dynamic response measurements. Towers 1 and 3, subjected to repeated strong-intensity earthquake ground motions, collapsed during the third repetition due to the accumulation of residual displacements. Tower 2 was subjected to a single near-fault ground motion representing an extreme event and collapsed during the test. Different collapse mechanisms were identified in the test program. Data collected from individual blocks showed in-plane and out-of-phase block rotation and sliding, which contributed to the system’s energy dissipation during the tests.

Abstract The 2023 M7.8 Kahramanmaraş/Pazarcik earthquake was larger and more destructive than what had been expected. Here we analyzed nearfield seismic records and developed a dynamic rupture model that reconciles different currently conflicting inversion results and reveals spatially non-uniform propagation speeds in this earthquake, with predominantly supershear speeds observed along the Narli fault and at the southwest (SW) end of the East Anatolian Fault (EAF). The model highlights the critical role of geometric complexity and heterogeneous frictional conditions in facilitating continued propagation and influencing rupture speed. We also constrained the conditions that allowed for the rupture to jump from the Narli fault to EAF and to generate the delayed backpropagating rupture towards the SW. Our findings have important implications for understanding earthquake hazards and guiding future response efforts and demonstrate the value of physics based dynamic modeling fused with near-field data in enhancing our understanding of earthquake mechanisms and improving risk assessment.

The 2023 M7.8 Kahramanmaraş/Pazarcik earthquake was larger and more destructive than what had been expected. Here we analyzed near-field seismic records and developed a dynamic rupture model that reconciles different currently conflicting inversion results and reveals spatially non-uniform propagation speeds in this earthquake, with predominantly supershear speeds observed along the Narli fault and at the southwest (SW) end of the East Anatolian Fault (EAF). The model highlights the critical role of geometric complexity and heterogeneous frictional conditions in facilitating continued propagation and influencing rupture speed. We also constrained the conditions that allowed for the rupture to jump from the Narli fault to EAF and to generate the delayed backpropagating rupture towards the SW. Our findings have important implications for understanding earthquake hazard and guiding future response efforts and demonstrates the value of physics-based dynamic modeling fused with near-field data in enhancing our understanding of earthquake mechanisms and improving risk assessment.

The 2023 M7.8 Kahramanmaraş/Pazarcik earthquake was larger and more destructive than what had been expected. Here we analyzed near-field seismic records and developed a dynamic rupture model that reconciles different currently conflicting inversion results and reveals spatially non-uniform propagation speeds in this earthquake, with predominantly supershear speeds observed along the Narli fault and at the southwest (SW) end of the East Anatolian Fault (EAF). The model highlights the critical role of geometric complexity and heterogeneous frictional conditions in facilitating continued propagation and influencing rupture speed. We also constrained the conditions that allowed for the rupture to jump from the Narli fault to EAF and to generate the delayed backpropagating rupture towards the SW. Our findings have important implications for understanding earthquake hazard and guiding future response efforts and demonstrates the value of physics-based dynamic modeling fused with near-field data in enhancing our understanding of earthquake mechanisms and improving risk assessment.

Abstract In this paper a modeling method is validated at multiple scales for the seismic performance of multiblock tower structure (MTS). MTS are a proposed concept for large‐capacity gravitational energy storage that will enable renewable energy sources. The structure modeled is a tower of 7144 nominally identical blocks arranged in a 38‐layered annular pattern with no adhesive mechanisms between the blocks or the blocks and the foundation. The level set discrete element method is used to model the dynamics of the tower structure experiencing a ground motion. Experimental determination of each model parameter is shown from the use of individual blocks before construction. Close comparisons to experimental results are shown for the dynamic motion of the tower over a full ground motion time history for multiple scales, materials and ground motions. When the tower was brought to failure, the two ground motions used produced distinct failure modes of the tower showing both a peeling and buckling behavior. Both the effect of the friction coefficient and unequal block heights are investigated. Friction coefficient has a noticeable effect on the amplitude of motion of the tower while the unevenness of the block heights affects mostly the structural speed.

The Mw7.8 Kahramanmaraş Earthquake was larger and more destructive than what had been expected for the tectonic setting in Southeastern Turkey. By using near-field records we provide evidence for early supershear transition on the splay fault that hosted the nucleation and early propagation of the first rupture that eventually transitioned into the East Anatolian fault. The two stations located furthest from the epicenter show a larger fault parallel particle velocity component relative to the fault normal particle velocity component; a unique signature of supershear ruptures that has been identified in theoretical and experimental models of intersonic rupture growth. The third station located closest to the epicenter, while mostly preserving the classical sub-Rayleigh characteristics, it also features a small supershear pulse clearly propagating ahead of the original sub-Rayleigh rupture. This record provides, for the first time ever, field observational evidence for the mechanism of intersonic transition. By using the two furthest stations we estimate the instantaneous supershear rupture propagation speed to be ~1.55 Cs and the sub-Rayleigh to supershear transition length to be around 19.45 km, very close to the location of the station nearest to the epicenter. This early supershear transition might have facilitated the continued propagation and triggering of slip on the nearby East Anatolian Fault leading to amplification of the hazard. The complex dynamics of the Kahramanmaraş earthquake warrants further studies.

Many large and damaging earthquakes on mature faults in the Earth’s crust propagate along layers of rock gouge, the fine granular material produced by comminution during sliding. Characterizing gouge rheology is of paramount importance to improve our understanding of earthquake physics, as friction controls key processes of earthquakes, including nucleation, propagation and arrest and how damaging they can be.  In this work, we characterize friction evolution in rock gouge layers during the propagation of dynamic ruptures in a laboratory setting. The experimental setup features a hybrid configuration with a specimen made of an analog material and a rock gouge layer embedded along the interface. This configuration allows us to trigger dynamic ruptures due to the lower shear modulus of the analogue material while at the same time study the gouge frictional behavior during spontaneously evolving dynamic events. Ruptures are captured by the use of digital image correlation coupled with ultrahigh-speed photography. Our measurements reveal dramatic friction variations, with the gouge layer initially displaying strengthening behavior and inhibiting earthquake rupture propagation. However, the gouge layer later features dramatic frictional strength losses, and hosts rupture re-nucleation enabled by dynamic stressing and marked friction weakening at higher slip velocities. Our measurements of the weakening and strengthening behavior of friction in fine rock gouge illustrate the strong dependence of their rheology on slip velocity and related processes, including shear heating, localization/delocalization of shear, and dilation/compaction of the granular shear layer.

Abstract This work examines the sub-shear and super-shear steady-state growth of mode III fractures in flexoelectric materials, nonetheless, exhibiting Mach type shock wave patterns that resemble reported lattice dynamics results and three-dimensional calculations and experiments. Our mathematical models provide weak discontinuous solutions of the steady-state dynamic equations. In flexoelectric solids, super-shear rupture is possible with Mach lines appearing at sub-shear as well as super-shear crack rupture velocities. This is contrary to classical singular elastodynamics, where the notions of super-shear growth and hyperbolicity coincide. The results show that the deformation near the crack-tip agrees with studies based on lattice dynamics. In the first part of this work, a novel finite element approach has been developed where the problem is decomposed into two prestressed plates that are interconnected, resulting into the predicted radiation patterns and Mach lines. The polarization field is obtained from the calculated displacement field and is used in turn to calculate the magnetic and the electric fields. The analysis offers an analogy to the co-seismic magnetic fields encountered during mode III dominated earthquake rupture events.

Abstract In our previous study (Part I), the anti-plane steady-state hyperbolic mode III fracture of a magneto-flexoelectric material was solved for the displacement, the polarization, and the magnetic fields. The solution, however, was based on the assumption of the development of strain discontinuities, and the propagation of the crack-tip was related to a critical shear strain. However, in the current study, the asymptotic details of the fields close to the crack-tip were investigated. The asymptotic analysis assumes strain continuity at the crack-tip (discontinuity in the strain gradients) and reveals the existence of a positive dynamic J-integral. The asymptotic analysis was performed not only for hyperbolic but also for elliptic conditions, and the energy release rate was calculated as a function of the crack-tip velocity in both regimes. These results are very different from those predicted by classical singular elastodynamics, where the dynamic J-integral is zero when super-shear is attained and there can be only an elliptic solution. Moreover, the results are very useful for couple-stress elastodynamics where equivalent length scales are present due to the analogy with flexoelectricity.

The Mw7.8 Kahramanmaraş Earthquake was larger and more destructive than what had been expected for the tectonic setting in Southeastern Turkey. By using near-field records we provide evidence for early supershear transition on the splay fault that hosted the nucleation and early propagation of the first rupture that eventually transitioned into the East Anatolian fault. We also find, for the first time ever, field observational evidence showing the mechanism of sub-Rayleigh to supershear transition. We estimate the instantaneous supershear rupture propagation speed to be ∼ 1.55Cs and the sub-Rayleigh to supershear transition length to be around ∼ 19.45 km, very close to the location of one of the stations, closest to the epicenter. This early supershear transition might have facilitated the continued propagation and triggering of slip on the nearby East Anatolian Fault leading to amplification of the hazard. The complex dynamics of the Kahramanmaras ̧ earthquake warrants further studies.

The Mw 7.8 Kahramanmaraş Earthquake was larger and more destructive than what had been expected for the tectonic setting in Southeastern Turkey. By using near-field records we provide evidence for early supershear transition on the splay fault that hosted the nucleation and early propagation of the first rupture that eventually transitioned into the East Anatolian fault. The two stations located furthest from the epicenter show a larger fault parallel particle velocity component relative to the fault normal particle velocity component; a unique signature of supershear ruptures that has been identified in theoretical and experimental models of intersonic rupture growth. The third station located closest to the epicenter, while mostly preserving the classical sub-Rayleigh characteristics, it also features a small supershear pulse clearly propagating ahead of the original sub-Rayleigh rupture. This record provides, for the first time ever, field observational evidence for the mechanism of intersonic transition. By using the two furthest stations we estimate the instantaneous supershear rupture propagation speed to be ~ 1.55 Cs and the sub-Rayleigh to supershear transition length to be around 19.45 km, very close to the location of the station nearest to the epicenter. This early supershear transition might have facilitated the continued propagation and triggering of slip on the nearby East Anatolian Fault leading to amplification of the hazard. The complex dynamics of the Kahramanmaraş earthquake warrants further studies.