Slip Pulses Research Articles

Doubly Stochastic Earthquake Source Model: “Omega-Square” Spectrum and Low High-Frequency Directivity Revealed by Numerical Experiments

Since its formulation in 1967–1970, the classical ω −2 model of earthquake source spectrum awaits a consistent theoretical foundation. To obtain one, stochastic elements are incorporated both into the final structure of the fault and into the mode of rupture propagation. The main components of the proposed “doubly stochastic” model are: (1) the Andrews’s concept, that local stress drop over a fault is a random self-similar field; (2) the concept of rupture with running slip pulse, after Heaton; (3) the hypothesis that a rupture front is a tortuous, multiply connected (“lacy”) fractal polyline that occupies a strip of finite width close to the slip-pulse width; and (4) the assumption that the propagation distance of fault-guided, mostly Rayleigh waves from a failing spot on a fault is determined by the slip-pulse width. Waveforms produced by this model are determined based on the fault asperity failure model after Das and Kostrov. Properties of the model are studied by numerical experiments. At high frequency, simulated source spectra behave as ω −2, and acceleration spectra are flat. Their level, at a given seismic moment and rms stress drop, is inversely related to the relative width of the slip pulse. When this width is relatively low, a well-defined second corner frequency (lower cutoff of acceleration spectrum) is seen. The model shows clear dependence of propagation-related directivity on frequency. Between the first and the second corner frequency, amplitude spectra are strongly enhanced for the forward direction; whereas, above the second corner frequency, directivity is significantly reduced. Still, it is not inhibited totally, suggesting incomplete incoherence of the simulated radiator at high frequencies.

Transonic gliding edge dislocations/slip pulse near and on an interface/fault

Three problems are solved in this paper that are related to transonic earthquakes. (1) The shear shock Mach wave emanating from a transonic gliding edge dislocation which impacts an interface. (2) A transonic edge dislocation gliding parallel and near to an interface. (3) The transonic edge dislocation gliding on the interface itself. The dislocation is in uniform or quasi‐uniform motion. The interface separates isotropic elastic material of slightly different properties. The first and last problems are essentially the same. The problems are solved with the building blocks of Mach waves, with logarithmic and arctan distributions of infinitesimal shock waves and with subsonic (with respect to longitudinal wave velocities) glide edge and climb edge dislocations. The need for the smeared infinitesimal shock wave distributions is an interesting feature of the solutions. Distributions of infinitesimal smeared Mach waves will exist wherever physical discontinuities exist near a transonic dislocation. A slip pulse of infinitesimal moving edge dislocations can be self sharpening because of attractive forces between like sign dislocations in certain velocity ranges.

Reproducing the supershear portion of the 2002 Denali earthquake rupture in laboratory

A notable feature of the 2002 Mw7.9 Denali, Alaska, earthquake was that a unique set of near-field seismic ground motion records, at Pump Station 10 (PS10), captured the passage of a supershear rupture followed by what was surmised to be a secondary slip pulse, ‘Trailing Rayleigh Pulse’ (Dunham and Archuleta, 2004; Mello et al., 2010). Motivated by the unique features contained in these near-field ground motion records, which were obtained only 3 km away from the fault, a series of scaled laboratory earthquake experiments was conducted in an attempt to replicate the dominant features of the PS10 ground motion signatures. Particle velocity records bearing a striking similarity to the Denali ground motion records are presented and discussed. The success of the comparison opens up the possibility of routinely generating near source ground motion records in a scaled and controlled laboratory setting that could be of great societal interest towards assessing seismic hazard from large and potentially devastating earthquakes.

The existence of a critical length scale in regularised friction

We study a regularisation of Coulomb's friction law on the propagation of local slip at an interface between a deformable and a rigid solid. This regularisation, which was proposed based on experimental observations, smooths the effect of a sudden jump in the contact pressure over a characteristic length scale. We apply it in numerical simulations in order to analyse its influence on the behaviour of local slip. We first show that mesh convergence in dynamic simulations is achieved without any numerical damping in the bulk and draw a convergence map with respect to the characteristic length of the friction regularisation. By varying this length scale on the example of a given slip event, we observe that there is a critical length below which the friction regularisation does not affect anymore the propagation of the interface rupture. A spectral analysis of the regularisation on a periodic variation of Coulomb's friction is conducted to confirm the existence of this critical length. The results indicate that if the characteristic length of the friction regularisation is smaller than the critical length, a slip event behaves as if it was governed by Coulomb's law. We therefore propose that there is a domain of influence of the friction regularisation depending on its characteristic length and on the frequency content of the local slip event. A byproduct of the analysis is related to the existence of a physical length scale characterising a given frictional interface. We establish that the experimental determination of this interface property may be achieved by experimentally monitoring slip pulses whose frequency content is rich enough.

Additional shear resistance from fault roughness and stress levels on geometrically complex faults

The majority of crustal faults host earthquakes when the ratio of average background shear stress τb to effective normal stress σeff is τb/σeff≈0.6. In contrast, mature plate‐boundary faults like the San Andreas Fault (SAF) operate at τb/σeff≈0.2. Dynamic weakening, the dramatic reduction in frictional resistance at coseismic slip velocities that is commonly observed in laboratory experiments, provides a leading explanation for low stress levels on mature faults. Strongly velocity‐weakening friction laws permit rupture propagation on flat faults above a critical stress level τpulse/σeff≈0.25. Provided that dynamic weakening is not restricted to mature faults, the higher stress levels on most faults are puzzling. In this work, we present a self‐consistent explanation for the relatively high stress levels on immature faults that is compatible with low coseismic frictional resistance, from dynamic weakening, for all faults. We appeal to differences in structural complexity with the premise that geometric irregularities introduce resistance to slip in addition to frictional resistance. This general idea is quantified for the special case of self‐similar fractal roughness of the fault surface. Natural faults have roughness characterized by amplitude‐to‐wavelength ratios α between 10−3 and 10−2. Through a second‐order boundary perturbation analysis of quasi‐static frictionless sliding across a band‐limited self‐similar interface in an ideally elastic solid, we demonstrate that roughness induces an additional shear resistance to slip, or roughness drag, given by τdrag=8π3α2G∗Δ/λmin, for G∗=G/(1−ν) with shear modulus Gand Poisson's ratio ν, slip Δ, and minimum roughness wavelength λmin. The influence of roughness drag on fault mechanics is verified through an extensive set of dynamic rupture simulations of earthquakes on strongly rate‐weakening fractal faults with elastic‐plastic off‐fault response. The simulations suggest that fault rupture, in the form of self‐healing slip pulses, becomes probable above a background stress level τb≈τpulse+τdrag. For the smoothest faults (α∼10−3), τdrag is negligible compared to frictional resistance, so that τb≈τpulse≈0.25σeff. However, on rougher faults (α∼10−2), roughness drag can exceed frictional resistance. We expect that τdrag ultimately departs from the predicted scaling when roughness‐induced stress perturbations activate pervasive off‐fault inelastic deformation, such that background stress saturates at a limit (τb≈0.6σeff) determined by the finite strength of the off‐fault material. We speculate that this strength, and not the much smaller dynamically weakened frictional strength, determines the stress levels at which the majority of faults operate.

Coseismic recrystallization during shallow earthquake slip

Solidified frictional melts, or pseudotachylytes, remain the only unambiguous indicator of seismic slip in the geological record. However, pseudotachylytes form at >5 km depth, and there are many rock types in which they do not form at all. We performed low- to high-velocity rock friction experiments designed to impose realistic coseismic slip pulses on calcite fault gouges, and report that localized dynamic recrystallization may be an easy-to-recognize microstructural indicator of seismic slip in shallow, otherwise brittle fault zones. Calcite gouges with starting grain size −1 , and total displacements between 1 and 4 m. At coseismic slip velocities ≥0.1 m s −1 , the gouges were cut by reflective principal slip surfaces lined by polygonal grains 2 + CaO. The recrystallized calcite aggregates resemble those found along the principal slip surface of the Garam thrust, South Korea, exhumed from

Dynamical model of faulting in two dimensions and self-healing of large fractures

We describe a model for the simulation of extended two-dimensional in-plane dynamical ruptures and for the rapid calculation of statistical properties of repeated model-seismicity events. The discretization involves first- and second-nearest neighbors and is isotropic in both compression and shear properties. All rupture events obey a fracture criterion in the appropriate coordinate frame and numerical oscillations in slip velocity at crack tips due to discretization are minimized. The rupture velocities of fractures, in cases of homogeneous stress drop equal to the strength, are the supershear P-wave velocity in the direction of the prestress and the S-wave velocity in the perpendicular direction. We use the model to study the growth and healing of individual faults to understand the formation of propagating slip pulses. We confirm two mechanisms for the generation of isolated rupture pulses that have been proposed, namely, (1) a decrease in the dynamical friction with accelerating slip and (2) the encounter of the growing crack with extended regions of large difference between the threshold fracture stress and the prestress. We describe a third mechanism which is that of a velocity-dependent friction that operates equally on both the phases of increasing and decreasing slip velocities and has a characteristic length scale. It is a proxy for energy loss by radiation in a three-dimensional medium. In the case of an elongated rectangular model fault with an upper free surface and lower rigid boundary, pulses develop due to the influence of stress waves reflected from the rigid bottom boundary. In general, the excess of strength over stress drop controls crack fracture speeds; if it is too large, the crack stops. Under homogeneous stress conditions, isolated slip pulses are controlled by the spatial distribution of heterogeneities and by the velocity-dependent friction parametrization.

Properties of inelastic yielding zones generated by in-plane dynamic ruptures—II. Detailed parameter-space study

We perform a detailed parameter-space study on properties of yielding zones generated by 2-D in-plane dynamic ruptures on a planar fault with different friction laws and parameters, different initial stress conditions, different rock cohesion values, and different contrasts of elasticity and mass density across the fault. The focus is on cases corresponding to large strike-slip faults having high angle (ψ = 45◦) to the maximum compressive background stress. The simulations and analytical scaling results show that for crack-like ruptures (1) the maximum yielding zone thickness T_(max) linearly increases with rupture distance L and the ratio Tmax/L is inversely proportional to (1 + S)^2 with S being the relative strength parameter; (2) the potency density e^p_0 decays logarithmically with fault normal distance at a rate depending on the stress state and S; (3) increasing rock cohesion reduces T_(max)/L, resulting in faster rupture speed and higher inclination angle Ф of expected microfractures on the extensional side of the fault. For slip pulses in quasi-steady state, T is approximately constant along strike with local values correlating with the maximum slip velocity (or final slip) at a location. For a bimaterial interface with ψ = 45◦, the energy dissipation to yielding contributes to developing macroscopically asymmetric rupture (at the scale of rupture length) with the same preferred propagation direction predicted for purely elastic cases with Coulomb friction. When ψ = 10◦, representative for thrust faulting, the energy dissipation to yielding leads to opposite preferred rupture propagation. In all cases, Ф is higher on average on the compliant side. For both crack and pulse ruptures with ψ = 45◦, T decreases and e^p_0 increases for conditions representing greater depth. Significant damage asymmetry of the type observed across several large strike-slip faults likely implies persistent macroscopic rupture asymmetry (unilateral cracks, unilateral pulses or asymmetric bilateral pulses). The results on various features of yielding zones expected from this and other studies are summarized in a table along with observations from the field and laboratory experiments.

Contrasting rupture processes during the April 11, 2010 deep-focus earthquake beneath Granada, Spain

The Mw=6.3 deep-focus earthquake beneath Granada, Spain, in 2010 consisted of three resolvable sub-events occurring within a time span of 5s. Estimated sub-event seismic moment partitioning is 12%, 7% and 81%, respectively. All sub-events had similar focal mechanisms with a vertical and a near-horizontal nodal plane, and all occurred within 5km of each other at a similar depth, suggesting rupture on the near-horizontal nodal plane. However, directivity analyses indicate that the first sub-event ruptured unilaterally on the vertical plane. Its modeled rupture length of ∼9km and stress drop of ∼2MPa are typical of crustal earthquakes. In contrast, the following sub-events show no clear directivity. The third, best resolved, sub-event had a hypocenter ∼2km from the first and a focal mechanism indistinguishable from the first, but it had a rupture dimension of <6.5km and a stress drop of >40MPa. This requires an ambient stress field significantly greater than the stress drop of the first sub-event, implying that the first sub-event ruptured as a slip pulse with a transient weakening mechanism. The large stress drops of the second and third sub-events suggest a crack-like rupture without fault healing and with nearly total stress drop. Fault-zone melting and metastable olivine are viable mechanisms for these ruptures. In contrast, the rupture characteristics of the first sub-event seem incompatible with most mechanisms currently under consideration for deep-focus earthquakes.

Depth‐dependent mode of tremor migration beneath Kii Peninsula, Nankai subduction zone

We investigated the migration mode of deep non‐volcanic tremor activity beneath Kii Peninsula, southwest Japan. Major tremor episodes are characterized by long‐term migration with a velocity of about 10 km/day, propagating along the strike of the subducting plate. Similar tremor migration in Cascadia is accompanied by reverse propagation at speeds on the order of 100 km/day and much faster slip‐parallel migration at speeds on the order of 1000 km/day. We systematically searched for migrating tremor with clear linearity in space and time. As a result, we found tremor migrations at speeds ranging from 1 to 60 km/hr depending on the along‐dip position in the tremor zone. The observed decrease in migration speed with increasing measurement time scale suggests that migration is controlled by a diffusion process. The along‐strike migration at lower speeds, including both forward and backward directions relative to the long‐term migration episode, is concentrated at the updip side of the tremor zone, whereas the faster slip‐parallel migration is distributed over the entire zone. The long‐term migration seems to consist of and be excited by the propagation of along‐strike creep at the updip part. The concentration of along‐strike migrating tremor sequences at the updip side may reflect the existence of abundant fluid that accumulates at the corner of the mantle wedge. The faster slip‐parallel migrations represent projections of along‐strike fluctuations in slip pulse propagation controlled by striations along the plate interface.

A new paradigm for simulating pulse-like ruptures: the pulse energy equation

We investigate the chaotic behaviour of slip pulses that propagate in a spring block slider model with velocity weakening friction by numerically solving a computationally intensive set of n coupled non-linear equations, where n is the number of blocks. We observe that the system evolves into a spatially heterogeneous pre-stress after the occurrence of a sufficient number of events. We observe that, although the spatiotemporal evolution of the amplitude of a slip pulse in a single event is surprisingly complex, the geometric description of the pulses is simple and self-similar with respect to the size of the pulse. This observation allows us to write an energy balance equation that describes the evolution of the pulse as it propagates through the known pre-stress. The equation predicts the evolution of individual ruptures and reduces the computational time dramatically. The long-time solution of the equation reveals its multiscale nature and its potential to match many of the long-time statistics of the original system, but with a much shorter computational time.

Seismic and aseismic slip pulses driven by thermal pressurization of pore fluid

There are several lines of evidence that suggest that thermal pressurization (TP) of pore fluid within a low‐permeability fault core may play the key role in the development of earthquake slip. To elucidate effects of TP on spontaneous fault slip, we consider solutions for a steadily propagating slip pulse on a fault with a constant sliding friction, the level of which may reflect other thermally‐activated processes at the rupture front (such as the flash heating on asperities). Upon arrival of the pulse front, essentially undrained‐adiabatic TP takes place during the initial slip acceleration from the locked state with a corresponding reduction of the fault strength. With passage of time, the diminishing rate of heating (due to the reduced fault strength) and increasing rate of hydrothermal diffusion from the shear zone offset TP and result in partial recovery of the strength, slip deceleration and eventual locking and healing of the slip. We show that the rupture speedvr decreases with thickness hof the principal shear zone. For lab‐constrained values of fault‐gouge parameters, the TP‐pulse solution predicts seismic (vr∼ km/s) slip on a millimeter‐to‐cm thin principal shear zone; and aseismic slip withvr ∼ 10 km/day and slip rates 1–2 orders above the plate rate on a relatively thick (h∼ 1 m) shear zone. These and other predictions of the TP‐pulse model are consistent with the independent sets of observational constraints for large crustal and subduction interplate earthquakes, and slow slip transients (North Cascadia), respectively. Locking of the slip soon after the diffusive transport of the heat and pore fluid becomes efficient significantly limits the maximum co‐seismic temperature rise to values well below previous theoretical estimates. As a result, the onset of macroscopic melting and some of thermal decomposition reactions, recently suggested to explain strong co‐seismic fault weakening, are precluded over much of the seismogenic zone.

Bowed-string multiphonics analyzed by use of impulse response and the Poisson summation formula

By carefully positioning the bow and a lightly touching finger on the string, the string spectrum can be conditioned to provide narrow bands of pronounced energy. This leaves the impression of multiple complex tones with the normal (Helmholtz) fundamental as the lowest pitch. The phenomenon is seen to be caused by two additional signal loops, one on each side of the finger, which through the repeating slip pattern get phase locked to the full loop of the fundamental. Within the nominal period, however, the slip pulses will not be uniform like they are during the production of a normal "harmonic" or "flageolet" but may vary considerably in shape, size, and timing. For each string, there is a large number of bow/finger combinations that bear the potential of producing such tones. There are also two classes, depending on whether the bow or the finger is situated closest to the bridge. Touching the string with the finger closest to the bridge will somewhat emphasize the (Helmholtz) fundamental. The technique is applicable to double bass and cello, while less practical on shorter-stringed instruments. Analyses based on impulse responses and the Poisson summation formula provide an explanation to the underlying system properties.

Earthquake Ruptures with Strongly Rate-Weakening Friction and Off-Fault Plasticity, Part 1: Planar Faults

Abstract Observations demonstrate that faults are fractal surfaces with deviations from planarity at all scales. We study dynamic rupture propagation on self-similar faults having root mean square (rms) height fluctuations of order 10 -3 to 10 -2 times the profile length. Our 2D plane strain models feature strongly rate-weakening fault friction and off-fault Drucker–Prager viscoplasticity. The latter bounds otherwise unreasonably large stress concentrations in the vicinity of bends. Our choice of a cohesionless yield function prevents tensile stress states and thus fault opening. A consequence of strongly rate-weakening friction is the existence of a critical background stress level above which self-sustaining rupture propagation, in the form of self-healing slip pulses, first becomes possible. Around this level, at which natural faults are expected to operate, ruptures become extremely sensitive to fault roughness and exhibit substantial fluctuations in rupture velocity. Except for shallow inclinations of the maximum compressive stress to the fault (less than about 20°), the fluctuations are anticorrelated with the local fault slope. These accelerations and decelerations of the rupture, together with naturally emerging slip heterogeneity, excite waves of all wavelengths and result in ground acceleration spectra that are flat at high frequency, consistent with observed strong motion records.

Periodicity, chaos and localization in a Burridge-Knopoff model of an earthquake with rate-and-state friction

SUMMARY We investigate the emergent dynamics when the slip law formulation of the non-linear rateand-state friction law is attached to a Burridge–Knopoff spring-block model. We derive both the discrete equations and the continuum formulation governing the system in this framework. The discrete system (ODEs) exhibits both periodic and chaotic motion, where the system’s transition to chaos is size-dependent, that is, how many blocks are considered. From the discrete model we derive the non-linear elastic wave equation by taking the continuum limit. This results in a non-linear partial differential equation (PDE) and we find that chaos ensues when the same parameter is increased. This critical parameter value needed for the onset of chaos in the continuous model is much smaller than the value needed in the case of a single block and we discuss the implications this has on dynamic modelling of earthquake rupture with this specific friction law. Most importantly, these results suggest that the friction law is scale-dependent, thus caution should be taken when attaching a friction law derived at laboratory scales to full-scale earthquake rupture models. Furthermore, we find solutions where the initial slip pulse propagates like a travelling wave, or remains localized in space, suggesting the presence of soliton and breather solutions. We discuss the significance of these pulse-like solutions and how they can be understood as a proxy for the propagation of the rupture front across the fault surface during an earthquake. We compute analytically the conditions for soliton solutions and by exploring the resulting parameter space, we introduce a possible method for determining a range of suitable parameter values to be used in future dynamic earthquake modelling.

Nucleation and Propagation of Quasi-Static Interfacial Slip Pulses

We report thorough experimental characterization of the nucleation and stationary propagation of self-healing slip pulses at a gel/glass interface under constant stress loading. We show that the slip front can be nucleated heterogeneously at the edges of the contact or homogeneously inside it. The resulting slip front can propagate both in the direction of slip or backwards; its velocity depends on the applied stress through a local tip process that is independent of specimen geometry. The backward tip velocity is slower than the front tip velocity by a factor of 4. Both velocities increase linearly with the applied stress jump, with a threshold below which no pulse can form. The pulse length and the total slip is found to be a linear function of the tip velocity. We address the complex problem of selection of the pulse velocity, length and associated slip by using a simplified analytical model. We emphasize the crucial role played by non-linearities of the friction law.

Episodic tremors and slip in Cascadia in the framework of the Frenkel-Kontorova model

Received 27 August 2010; revised 20 October 2010; accepted 28 October 2010; published 13 January 2011. [1] The seismic moment for regular earthquakes is proportional to the cube of rupture time. A second class of phenomena, collectively called slow earthquakes, has very different scaling. We propose a model, inspired from the phenomenology of dislocation dynamics in crystals, that is consistent with the scaling relations observed in the Cascadia episodic tremor and slip (ETS) events. Two fundamental features of ETS are periodicity and migration. In the northern Cascadia subduction zone, ETS events appear every 14.5 months or so. During these events, tremors migrate along‐strike with a velocity of 10 km/day and simultaneously zip back and forth in the relative plate‐ motion direction with at ypical velocity of 50 km/h. Our model predicts the formation of a sequence of slip pulses on the boundary of the plates, which describes the major features of fault dynamics, including periodicity and the migration pattern of tremors. Citation: Gershenzon, N. I., G. Bambakidis, E. Hauser, A. Ghosh, and K. C. Creager (2011), Episodic tremors and slip in Cascadia in the framework of the Frenkel‐Kontorova model, Geophys. Res. Lett., 38, L01309,

A numerical study on stick–slip motion of a brake pad in steady sliding

A numerical model for an elastic brake pad sliding under constant load and with constant velocity over a rigid surface is investigated by finite element analysis. The geometry is taken to be two-dimensional, the contact is assumed to follow the laws of continuum mechanics and temporal and spatial resolution are such that dynamical effects localized at the interface are resolved. It turns out that at the contact interface localized slip events occur either in the form of long-lasting slip pulses, or in the form of brief local relaxations. Macroscopically steady sliding, macroscopic stick–slip motion or slip–separation dynamics occurs, depending on the macroscopic relative velocity. While structural oscillations of the brake pad do not seem to play a significant role during steady sliding at least one structural oscillation mode becomes synchronized with the interfacial dynamics during stick–slip or slip–separation motion. Assuming a given friction law for the interface, the macroscopically observed friction coefficient depends considerably on the underlying dynamics on the interface.

On the relations between fracture energy and physical observables in dynamic earthquake models

We explore the relationships between the fracture energy density (EG) and the key parameters characterizing earthquake sources, such as the rupture velocity (vr), the total fault slip (utot), and the dynamic stress drop (Δτd). We perform several numerical experiments of three‐dimensional, spontaneous, fully dynamic ruptures developing on planar faults of finite width, obeying different governing laws and accounting for both homogeneous and heterogeneous friction. Our results indicate that EG behaves differently, depending on the adopted governing law and mainly on the rupture mode (pulselike or cracklike, sub‐ or supershear regime). Subshear, homogeneous ruptures show a general agreement with the theoretical prediction of EG ∝ , but for ruptures that accelerate up to supershear speeds it is difficult to infer a clear dependence of fracture energy density on rupture speed, especially in heterogeneous configurations. We see that slip pulses noticeably agree with the theoretical prediction of EG ∝ utot2, contrarily to cracklike solutions, both sub‐ and supershear and both homogeneous and heterogeneous, which is in agreement with seismological inferences, showing a scaling exponent roughly equal to 1. We also found that the proportionality between EG and Δτd2, expected from theoretical predictions, is somehow verified only in the case of subshear, homogeneous ruptures with RD law. Our spontaneous rupture models confirm that the total fracture energy (the integral of EG over the whole fault surface) has a power law dependence on the seismic moment, with an exponent nearly equal to 1.13, in general agreement with kinematic inferences of previous studies. Overall, our results support the idea that EG should not be regarded as an intrinsic material property.

Pulse‐like dynamic earthquake rupture propagation under rate‐, state‐ and temperature‐dependent friction

Healing of faults is an important process in earthquake source physics since it accounts for a rapid restrengthening of the fault traction and for a consequent short slip duration, as indicated by slip inversions of seismic data. In this paper we show that a laboratory‐derived constitutive model, with an explicit dependence on the temperature developed by frictional heat, can provide a suitable explanation for the generation of self‐healing slip pulses. The model requires neither special modifications at low or high speeds nor the introduction of heterogeneities in the material properties, as previously proposed. We also demonstrate through numerical experiments of 3‐D ruptures that the temperature evolution can discriminate between crack‐like and slip pulses mode of propagation. In particular, we find that for a moderate level of strain localization (slipping zone width larger than 20 mm) ruptures behave as classical enlarging cracks.