Earthquake cycles and shear zones: interplay between earthquakes, aseismic fault slip, and bulk viscous deformation

<p>The interaction between the seismogenic portion of faults and their ductile roots is central to understanding the mechanics of seismic cycles. It is well established that faults are highly localized within the cold and brittle upper crust, but less is known about fault and shear zone structure in the warmer, more ductile, lower crust and in the upper mantle. Increasing temperature with depth causes two transitions in behavior: a frictional transition from seismic to aseismic fault behavior and a transition from brittle to ductile off-fault deformation (BDT). To explore the effects of these two transitions on seismic cycle characteristics (e.g., recurrence interval, nucleation depth, and down-dip limit of coseismic rupture), we simulate seismic cycles on a 2D strike-slip fault. All phases of the earthquake cycle are simulated, allowing the model to spontaneously generate earthquakes and to capture aseismic fault slip and off-fault viscous flow in the interseismic period. The fault is represented with rate-and-state friction. In the off-fault material, distributed viscous flow occurs through dislocation creep. We also consider two possible weakening mechanisms that may be active in lower crustal shear zones: shear heating and grain size reduction, which changes the ductile rheology from dislocation to diffusion creep. This model makes it possible to self-consistently simulate the variations of stress, strain rate, and grain size in the vicinity of a strike-slip fault.</p><p>We find that the viscous shear zone beneath the fault (defined as the region of elevated viscous strain rate) is roughly elliptically shaped, extending up to 10 km below the fault and with a width of 1 to 3 km. When weakening mechanisms are neglected, the BDT occurs below the depth of the transition from seismic to aseismic fault slip. In these cases, seismic cycle characteristics are similar to those of a traditional elastic cycle simulation that neglects viscoelastic deformation. However, the inclusion of shear heating, which produces a thermal anomaly relative to the background geotherm, shallows the BDT enough to limit the down-dip propagation of coseismic slip in some cases. In these cases, earthquakes penetrate 1-2 km into the shear zone, consistent with observations of zones in which both viscous flow and coseismic slip occur. Also, in these simulations, very little aseismic fault slip occurs. Instead, tectonic plate motion is accommodated primarily through coseismic slip and bulk viscous flow. Preliminary simulations that include the effects of grain size reduction within the shear zone show similar effects. Both weakening mechanisms narrow the shear zone by up to 20%, suggesting that the fault also plays a large role in controlling shear zone localization.</p>

Coseismic displacement interferences and patterns along subduction megathrusts: insights from analog modelling

<p>Finding a deformation pattern that is representative of a given stage of the seismic cycle of subduction megathrusts is crucial as this might provide clues about the upcoming earthquake. Here we focus on the short term interaction between seismic asperities and in particular on how geodetic velocities change in response to ruptures of an along-strike neighbor portion of the megathrust. Enhanced megathrust coupling, slab acceleration, in plane bending of the overriding plate, continental-scale viscoelastic mantle relaxation have been proposed as potentially responsible driving mechanisms. However, the paucity of observations from natural cases and the multiple- interrelated contributions that act at different spatial and temporal scales complicate the understanding of this process.</p><p>We use an analog model that simulates a series of laterally partial ruptures and analyze systematically the effect of slip episodes on deformation history of the neighbor “receiver” region. The analog model has the advantage of reproducing tens of seismic cycles with well controlled boundary conditions. The model shows that the deformation pattern associated to slip episodes has a characteristic twisting about a vertical axis. Such twisting interfere positively (causing velocity increase) or negatively (causing velocity decrease) with local interseismic velocity field depending on time since the last earthquake. Identifying accelerating or decelerating velocities in geodetic timeseries could be therefore informative of the seismic evolution of a subduction zone.</p>

Monitoring the dynamics of seismicity within the Kemin-Chilik zone, generating M≥8 earthquakes

The dynamics of seismic processes at the junction of the Tien Shan mountain building area and the Kazakh shield is presented in the paper. It is noted that the Tien Shan’s lithosphere over thrusts the Kazakh Shield’s lithosphere, and the Kazakh Shield’s lithosphere under thrusts beneath the Tien Shan’s lithosphere based on the seismic tomographic and seismotectonic data. Low-velocity heterogeneity is distinguished at the junction of these lithospheres, under where a low-velocity anomaly flow is assumed in the mantle. Marginal (active structures of the Ili basin, Zaili mountain range), and middle (active structures of the Kemin, Chilik basins, Kungei mountain range) subzones with characteristic seismicity and seismic regimes are formed here. Seismogenic zones are distinguished (from north to south): Predzaili, Kemin-Chilik, Predkungei. Powerful earthquakes with M>8 occur in the Kemin-Chilik seismogenic zone (about 250 km long and up to 25 km wide), and earthquakes with M=7-8 - in the Predza-ili and Predkungei seismogenic zones. The dynamics of the earthquakes’ sequence is predetermined by the dynamics of the hierarchy of faults and blocks in the junction zone. The sequence of earthquakes is expressed by the hierarchy of seismic cycles. Seismic activation period, a peak of seismic activation, a period of seismic activation’s decay, and seismic calm period are distinguished in every cycle. Strong earthquakes take place in a first-order cycle with a long period, significant and small earthquakes - in cycles with corresponding short periods. The seismicity level of the study area is determined by the trajectory of the seismic cycles’ association. Dynamic segmentation and dynamic sectorization, vectors of seismic activity directed from the east and west to the highly compressed central part of the region are noted in the spatial and temporal distribution of earthquakes at the junction of the Tien Shan and the Kazakh shield.

The Altotiberina Low-Angle Normal Fault (Italy) Can Fail in Moderate-Magnitude Earthquakes as a Result of Stress Transfer from Stable Creeping Fault Area

Geological and geophysical evidence suggests that the Altotiberina low-angle (dip angle of 15–20 ° ) normal fault is active in the Umbria–Marche sector of the Northern Apennine thrust belt (Italy). The fault plane is 70 km long and 40 km wide, larger and hence potentially more destructive than the faults that generated the last major earthquakes in Italy. However, the seismic potential associated with the Altotiberina fault is strongly debated. In fact, the mechanical behavior of this fault is complex, characterized by locked fault patches with a potentially seismic behavior surrounded by aseismic creeping areas. No historical moderate (5 ≤ Mw ≤ 5.9) nor strong (6 ≤ Mw ≤ 6.9)-magnitude earthquakes are unambiguously associated with the Altotiberina fault; however, microseismicity is scattered below 5 km within the fault zone. Here we provide mechanical evidence for the potential activation of the Altotiberina fault in moderate-magnitude earthquakes due to stress transfer from creeping fault areas to locked fault patches. The tectonic extension in the Umbria–Marche crustal sector of the Northern Apennines is simulated by a geomechanical numerical model that includes slip events along the Altotiberina and its main seismic antithetic fault, the Gubbio fault. The seismic cycles on the fault planes are simulated by assuming rate-and-state friction. The spatial variation of the frictional parameters is obtained by combining the interseismic coupling degree of the Altotiberina fault with friction laboratory measurements on samples from the Zuccale low- angle normal fault located in the Elba island (Italy), considered an older exhumed analogue of Altotiberina fault. This work contributes a better estimate of the seismic potential associated with the Altotiberina fault and, more generally, to low-angle normal faults with mixed-mode slip behavior.

The effect of multiple splay fault rupture on tsunamis

<p>Earthquake rupture on splay faults in subduction zones could pose a significant tsunami hazard, as they could accommodate more vertical displacement and are situated closer to the coast. To better understand this tsunami hazard, we model splay fault rupture dynamics and tsunami propagation and inundation constrained by a geodynamic seismic cycle (SC) model; building on work presented in Van Zelst et al. (2019). This two-dimensional modelling framework considers geodynamics, seismic cycles, dynamic ruptures, and tsunamis together for the first time. The SC model provides six blind splay fault geometries, self-consistent stress and strength conditions, and heterogeneous material properties in the domain. We find that all six splay faults are activated when the megathrust ruptures. The largest splay fault closest to the nucleation region ruptures immediately when the main rupture front passes the branching point. The other splay faults are activated through dynamic stress transfer from the main megathrust rupture or reflected waves from the surface. Splay fault rupture results in distinct peaks in the vertical surface displacements with a smaller wavelength and larger amplitudes. The effect of the vertical surface displacements also translates into the resulting tsunami, which consists of one large wave for the megathrust-only model and seven waves for the model including splay faults. Here, six of the waves can be attributed to the splay faults and the seventh wave results from the shallow tip of the megathrust. The waves from the rupture including splay faults have larger amplitudes and result in two episodes of coastal flooding. The first episode is due to the large wave caused by rupture on the largest splay fault nearest to the coast. The second flooding episode results from the combination and interference of the waves caused by the rest of the splay faults and the shallow megathrust tip. In contrast, the tsunami caused by rupture on only the megathrust has only one episode of flooding. Our results suggest that larger-than-expected tsunamis could be attributed to rupture on large splay faults. When multiple smaller splay faults rupture their effect on the tsunami might be hard to distinguish from a pure megathrust rupture. Considering the significant effects splay fault rupture can have on a tsunami, it is important to understand splay fault activation and to consider them in hazard assessment.</p><p>References:</p><p>Van Zelst, I., Wollherr, S., Madden, E. H. , Gabriel, A.-A., and Van Dinther, Y. (2019). Modeling megathrust earthquakes across scales: one-way coupling from geodynamics and seismic cycles to dynamic rupture. Journal of Geophysical Research: Solid Earth, 124, https://doi.org/10.1029/2019JB017539</p><p></p>

Predictability of large subduction earthquakes: insights from analog models and machine learning

<p>Despite the growing spatio-temporal density of geophysical observations, our understanding of the megathrust earthquake cycle continues to be limited by a series of factors, in particular the short observation time compared to mega-earthquake recurrence and the partial spatial coverage of geodetic data. Here, we attempt to compensate for these natural limitations by simulating dozens of seismic cycles in a laboratory-scale analogue model of subduction. The model creates analog earthquakes of magnitude Mw 6.2–8.3, with a coefficient of variation in recurrence intervals of 0.5, similar to real subduction megathrusts. Using a digital image correlation technique, we measure coseismic and interseismic deformation – this is akin to having a dense continuous geodetic network homogeneously distributed over the whole margin. We show how, by deciphering the spatially and temporally complex surface deformation history, machine learning can predict the timing and size of analog earthquakes. Then, we investigate data characteristics that maximize the performance of a machine learning binary classifier predicting slip-events imminence. We show how this framing can be used for designing an efficient geodetic network, and defining the minimum space-time coverage requirements for analog earthquake prediction. Converting the laboratory scale to the natural scale, we found that a 70-85 km wide coastal swath gives the most important information on slip imminence and that model performance is mainly 
influenced by the alarm duration, with density of stations and record length playing a secondary role. Under optimal monitoring conditions, about ten seismic cycles long record is enough to predict alarm periods in good agreement with those observed.</p>

Improving the Co-seismic Slip Distributions of Synthetic Catalogs With Real Observations

<p>Subduction zones are the most seismically active regions on the globe and about 90% of historical events, including the largest ones with the magnitude M>9, occurred along these regions (Hayes et al., 2018). Most of these events were followed by devastating tsunamis with, in some cases, perhaps unexpected wave height distributions. Observation of events in the megathrust environment reveals that some earthquakes are characterized by slip concentration on the very shallow part of the subduction zone. This shallow slip phenomenon was repeatedly observed in the last two decades for both ordinary megathrust events (e.g. 2010 Maule and 2011 Tohoku) and tsunami earthquakes (2006 Java and 2010 Mentawai). Shallow ruptures feature depleted short–period energy release and very slow rupture velocity possibly due to the presence of (hydrated) sediments (Lay et al., 2011; Lay 2014; Polet and Kanamori, 2000). Associated long rupture durations have been explained with fault mechanics-related rigidity and stress drop variation with depth (Bilek and Lay, 1999) or, more recently, with lower rigidity of surrounding materials (Sallares and Ranero, 2019).</p><p>The characteristics of co-seismic slip distribution have an important impact on tsunami hazard. There are numerous methods that have been proposed to generate stochastic slip distributions, also including shallow slip amplification (Le Veque et al., 2016; Sepulveda et al., 2017; Scala et al., 2019). However, these models need to be calibrated against slip models estimated for real events.</p><p>Here, we investigate similarities and differences between the synthetic slip distributions provided by Scala et al. (2019) and a suite of 144 slip models of real events that occurred in different subduction zones (Ye et al.,2016). In particular, Scala et al. (2019) model features shallow slip amplification in single events, whose relative probabilities are balanced to restore cumulative slip homogeneity on the fault plane over multiple seismic cycles. This study also aims to improve and/or calibrate this model to account for the behavior observed from real events.</p>

Resolving 3D coseismic deformation of the 2019 Mw 7.1 Ridgecrest earthquake using radar and optical data

<p>Measurements of surface displacement have been used in order to learn about seismic cycles, volcanoes, and other tectonic and non-tectonic processes. Ideally, the requirements to obtain useful measurements associated with seismic cycles are related to having a good spatial and temporal resolution, as surface deformation can occur in expected and unexpected faults, and in time intervals which vary from seconds (e.g. earthquake) to hundreds of years or even more (interseismic deformation).</p><p>Nowadays, satellite imagery provided by Synthetic Aperture Radar (SAR) or optical satellites fulfills those two aspects. Satellite images can cover large areas so that the fault rupture can be partially or totally visible. The problem of the radar technique is that for large earthquakes with surface rupture it cannot provide displacement maps in the near-field of the fault due to the large displacement gradient which causes phase decorrelation. Moreover, it is less sensitive to the horizontal displacement than vertical displacement. On the other hand, the main advantage of radar observing technique over the optical one is that the waves, emitted from a pulse-generating device, propagate through the atmosphere with almost no signal loss. This means that radar techniques operate under all weather conditions. Additionally, radar sensors are active, providing their own energy source, while optical are passive sensors that depend on external energy sources<!-- Here I would write also one sentence about the advantage of optical data, followed on by a sentence that the combination of optical and radar helps to retrieve 3D displacement field -->. Considering the benefits and the drawbacks of both sensing techniques, the opportunity of combining them helps the determination of a three-dimensional displacement field, illustrating a complete map of a seismic event.</p><p><!-- No proper here -->In consequence, the objective of this study is to provide a methodology, using radar (Sentinel-1) and optical (Sentinel-2) data, that leads to the determination of the three-dimensional displacement field associated with the 7<sup>th</sup> of July 2019, M<sub>w</sub> 7.1 Ridgecrest earthquake. The interferometric and offset tracking processing were computed using SNAP and GAMMA software, respectively, and for ascending and descending tracks products. For the optical data, cross-correlation using MicMac software was applied so that the displacement in the same area of interest was also derived. After obtaining the displacement for radar and optical data independently, a Least Square Adjustment (LSA) allowed to properly combine them considering the associated weight of each observation and finally compute the three-dimensional decomposition. Finally, it was possible to have a fully covered ground displacement measured from radar and optical sensors, and to better analyze the behavior of the tectonics in the area of study.</p>