David D. Oglesby

The Three-Dimensional Dynamics of Dipping Faults

Recent two-dimensional dynamic simulations of dip-slip faulting (Niel- sen, 1998; Oglesby et al., 1998, 2000; Shi et al., 1998) have shown that the asym- metric geometry of dip-slip faults that intersect the free surface can have large effects on the dynamics of earthquake rupture. The nonvertical dip angle of such faults leads to larger motion on the footwall than the hanging wall, as well as much larger motion from thrust/reverse faults than from normal faults with the same geometry and stress magnitudes. In the present work we perform full three-dimensional simulations of thrust/reverse, normal, and strike-slip faults, and show that the same effects exist in three dimensions. Strike-slip fault motion is either in between or lower than the motion of both dip-slip faults. Additional three-dimensional effects include strong rake rotation at the free surface. The results confirm the findings of the previous studies and further elucidate the dynamic effects of the free surface on fault rupture, slip, and ground motion. They are also borne out by early analyses of the 1999 Chi- Chi (Taiwan) thrust earthquake, which displayed higher motion on the hanging wall than on the footwall, and a strong oblique component of motion at the surface.

Heterogeneous fault stresses from previous earthquakes and the effect on dynamics of parallel strike‐slip faults

[1] We combine a viscoelastic model for the interseismic process and an elastodynamic model for the coseismic process to explore the dynamics (over multiple earthquake cycles) of two parallel strike-slip faults embedded in a two-dimensional full space. The step over fault geometry results in a buildup of heterogeneous fault stress near the step over. This heterogeneous stress accumulates at the early stage of the evolution of the fault system, and finally stabilizes after a number of earthquake cycles. The heterogeneity in fault stress varies with the geometrical parameters (e.g., width and along-strike overlap/ gap) of the step over, as well as the rupture history of the fault system. This heterogeneous fault stress from previous earthquakes has significant effects on earthquake rupture initiation, propagation, and termination. The locations with a low normal stress level near a step over are favorable points for earthquake initiation. Rupture can jump a 4 km wide compressional step over and a 8 km or wider dilational step over if the fault system has historically experienced many earthquakes. A young step over with less induced heterogeneity allows rupture to jump only smaller step over widths. These results may have important implications for seismic hazard analysis in areas where segmented strikeslip faults predominate, particularly for estimating maximum earthquake potential. Citation: Duan, B., and D. D. Oglesby (2006), Heterogeneous fault stresses from previous earthquakes and the effect on dynamics of parallel strike-slip faults, J. Geophys. Res., 111, B05309, doi:10.1029/2005JB004138.

Postseismic Fault Healing on the Rupture Zone of the 1999 M 7.1 Hector Mine, California, Earthquake

We probed the rupture zone of the October 1999 M 7.1 Hector Mine earthquake using repeated near-surface explosions in October 2000 and November 2001. Three dense linear seismic arrays were deployed across the north and south Lavic Lake faults (LLFs) that broke to the surface in the mainshock and across the Bullion fault (BF) that experienced minor slip in that event. Two explosions each year were detonated in the rupture zone, one on the middle and one on the south LLF. We found that P and S velocities of fault-zone rocks increased by 0.7%-1.4% and 0.5%-1.0% between 2000 and 2001, respectively. In contrast, the velocities for P and S waves in surrounding rocks increased much less. This trend indicates that the Hector Mine rupture zone has been healing by strengthening after the main- shock, most likely due to the closure of cracks that opened during the 1999 earth- quake. The observed fault-zone strength recovery is consistent with an apparent crack density decrease of 1.5% within the rupture zone. The ratio of travel-time decrease for P to S waves was 0.72, consistent with partially fluid-filled cracks near the fault zone. This restrengthening is similar to that observed after the 1992 M 7.4 Landers earthquake, which occurred 25 km to the west (Li and Vidale, 2001). We also find that the velocity increase with time varies from one fault segment to another at the Hector Mine rupture zone. We see greater changes on the LLFs than on the BF, and the greatest change is on the middle LLF at shallow depth. We tentatively conclude that greater damage was inflicted, and thus greater healing is observed, in regions with larger slip in the mainshock.

The Dynamics of Strike-Slip Step-Overs with Linking Dip-Slip Faults

Fault step-overs with linking dip-slip faults are common features on long strike-slip fault systems worldwide. It has been noted by various researchers that under some circumstances, earthquakes can jump across fault step-overs to cascade into large events, while under other circumstances rupture is arrested at step-overs. There is also evidence that fault step-overs may be preferential locations for earthquake nucleation. The present work uses the 3D finite element method to model the dynamics of strike-slip fault systems with step-overs and linking dip-slip faults. I find that the presence of a linking normal or thrust fault greatly increases the ability of earthquake rupture to propagate across the step-over, leading to a larger event. Additionally, dilational step-overs with linking normal faults are more prone to through-going rupture than compressional step-overs with linking thrust faults. This difference is due to the sign of the normal stress increment on the dip-slip fault caused by slip on the strike-slip segments: Slip on the strike-slip segments causes a negative (unclamping) normal stress increment on the linking normal fault in a dilational step-over, whereas the opposite effect occurs on the linking thrust fault in a compressional step-over. Even in cases for which both dilational and compressional step-overs can experience through-going rupture, dilational step-overs typically experience higher slip, particularly on the linking normal fault. In the compressional case, rupture nucleation on the linking thrust fault may increase the likelihood of through-going rupture compared to nucleation on one of the strike-slip segments. Near the intersections between the fault segments, the stress interaction between the fault segments also causes a significant rotation of rake away from that which would be inferred from the regional stress field. The results help to emphasize the importance of two-way interactions between nearby fault segments during the earthquake rupture process. The results also may have implications for the probability of large earthquakes along geometrically complex strike-slip fault systems, and may help explain why step-overs sometimes act as barriers and other times as nucleation locations for large earthquakes.

Fault Geometry and the Dynamics of the 1999 Chi-Chi (Taiwan) Earthquake

The 1999 M 7.6 Chi-Chi (Taiwan) earthquake produced a data set of unparalleled size and quality, particularly in the near-source region where data have been previously quite scarce. The large amount of near-source data allows the verification of many predictions of thrust-fault behavior for faults that intersect the surface of the earth. Through rigorous three-dimensional dynamic models of the Chi-Chi earthquake, it can be shown that many aspects of the observed near-source ground motion in this event are direct effects of the asymmetrical dipping fault geometry. These effects include the hanging wall moving more than the footwall (with strongly peaked velocities right at the fault trace) and a transition from predominantly thrust motion in the south of the fault to largely left-lateral motion in the north. Building on the work of Oglesby and Day (2001), the current work helps to delineate the effects of fault geometry, nonuniform prestress, and dynamic waves on the physics of the Chi-Chi earthquake and dip-slip faults in general. In particular, we find that a completely homogeneous prestress pattern still fits the gross features of the near-source ground motion quite well. Additionally, the strike-slip component of motion near the fault trace is seen to be a combination of dynamic and static effects. Finally, dynamic overshoot is seen to be much larger for dip-slip faults than for otherwise identical vertical faults. The results emphasize the necessity of rigorous models that correctly account for both the effects of fault geometry and dynamic waves in the rupture and slip processes. Manuscript received 31 July 2000.

Stochastic Fault Stress: Implications for Fault Dynamics and Ground Motion

Up to the present time, ground-motion calculations for future earthquakes have almost exclusively been made from kinematic models. However, dynamic faulting models offer many benefits over kinematic models, including the assurance the faulting models obey at least the general rules of elastodynamics and friction and contain a natural relationship between stress drop, slip, rise time, and rupture velocity. Dynamic models also offer insight into the physics of the rupture and slip processes and can show how these processes lead to patterns of ground motion. In the current work, we use the 3D finite-difference method and a suite of stochastic stress patterns, with variable assumptions on strength and stress inhomogeneity, to investigate two issues: (1) the effect of assumptions about stress pattern on the evolution of rupture and slip on the fault, and (2) the effect of these assumptions on the resultant ground motion. We find that stress drop has a complicated relationship with slip, rise time, and rupture velocity, especially in faults with strongly heterogeneous strength. We also find that these inhomogeneous-strength faults can produce highly inhomogeneous slip, even without any form of frictional restrengthening at healing time. Finally, we find that smoother strength models produce a better fit than the coarse models to the directivity pulse often observed on the surface. The results help to shed light on the transition of the faulting system between locally controlled and more globally controlled rupture and also show how dynamic models may be used to generate ground-motion estimates for seismic hazard calculations.

Dynamics of dip‐slip faulting: Explorations in two dimensions

Dynamic models of earthquake rupture and slip are a powerful method by which to investigate the physics of earthquakes. Owing to both conceptual and computational constraints, dynamic earthquake models have largely been limited to cases with geometrical symmetry, such as faults in unbounded media or vertical faults. However, there are both observational and theoretical reasons to believe that nonvertical dip-slip faults behave differently from faults with more symmetrical geometries. Previous observations have shown greater ground motion from thrust/reverse faults than normal faults and higher ground motion on hanging walls than on footwalls. In the present work, two-dimensional dynamic simulations of thrust/reverse and normal earthquakes show precisely these effects and also elucidate their causes. For typical nonvertical dip-slip faults the breakdown of symmetry with respect to the free surface allows radiated seismic waves to reflect off the free surface and to hit the fault again, altering the stress field on the fault. This process can lead to time-dependent normal stress and a feedback between the friction/rupture processes and seismic radiation. This interaction leads to thrust/reverse faults producing much higher fault and ground motion than normal faults with the same geometry and stress magnitudes. The asymmetric geometry also directly leads to higher motion on the hanging walls of such faults than on the footwalls. Simulations show that these effects occur for a variety of dip angles but only for faults that either intersect or closely approach the free surface. The results emphasize the strong effect that the free surface can have on the dynamics of fault rupture and slip.

Inverse Kinematic and Forward Dynamic Models of the 2002 Denali Fault Earthquake, Alaska

We perform inverse kinematic and forward dynamic models of the M 7.9 2002 Denali fault, Alaska, earthquake to shed light on the rupture process and dynamics of this event, which took place on a geometrically complex fault system in central Alaska. We use a combination of local seismic and Global Positioning System (gps) data for our kinematic inversion and find that the slip distribution of this event is characterized by three major asperities on the Denali fault. The rupture nucleated on the Susitna Glacier thrust fault, and after a pause, propagated onto the strike-slip Denali fault. Approximately 216 km to the east, the rupture abandoned the Denali fault in favor of the more southwesterly directed Totschunda fault. Three-dimensional dynamic models of this event indicate that the abandonment of the Denali fault for the Totschunda fault can be explained by the Totschunda fault’s more favorable orientation with respect to the local stress field. However, a uniform tectonic stress field cannot explain the complex slip pattern in this event. We also find that our dynamic models predict discontinuous rupture from the Denali to Totschunda fault segments. Such discontinuous rupture helps to qualitatively improve our kinematic inverse models. Two principal implications of our study are (1) a combination of inverse and forward modeling can bring insight into earthquake processes that are not possible with either technique alone, and (2) the stress field on geometrically complex fault systems is most likely not due to a uniform tectonic stress field that is resolved onto fault segments of different orientations; rather, other forms of stress heterogeneity must be invoked to explain the observed slip patterns.

Evidence from coseismic slip gradients for dynamic control on rupture propagation and arrest through stepovers

Received 30 July 2008; revised 14 November 2008; accepted 18 December 2008; published 27 February 2009. [1] Analysis of historic slip distributions from large-magnitude continental strike-slip earthquake ruptures reveals a pronounced correlation between the distance over which slip decreases as a rupture approaches a structural ‘‘step’’ in the fault and the ability of the rupture to propagate through the step. Our analysis of coseismic slip gradients near these stepovers indicates that in earthquakes in which slip decreases gradually toward a step, rupture will not continue on the next segment. Conversely, in earthquakes in which slip decreases abruptly, rupture commonly renucleates on the next segment. The fact that ruptures that stopped had low slip gradients near stepovers, relative to those that continued, indicates that rupture dynamics control the propagation of rupture through stepovers. These results corroborate dynamic rupture models that show that ruptures in which slip decreases abruptly at a step generate strong seismic waves that serve to renucleate rupture on the opposite side of the structural step. There are several potential causes for gradual decrease of displacement as rupture approaches a stepover, including transference of deformation onto subsidiary structures, rheological contrasts that could dictate favored rupture propagation directions, the existence of stress shadows from previous earthquakes within the system, or other material or frictional heterogeneities. All of these are potentially observable, and, if mapped systematically, could provide the basis for a strong constraint on the likely end points of future earthquakes. Inasmuch as earthquake magnitude is strongly dependent on the size of the rupture, such predictions would be of great utility as a basic component of scenario-based earthquake rupture forecasts.

Rupture Termination and Jump on Parallel Offset Faults

Proper evaluation of seismic hazard depends on accurate estimates of potential earthquake size. In areas with complex, multisegment fault systems, such an estimate in turn depends on an ability to predict the circumstances under which rupture may jump between fault segments. Many observational and numerical studies have analyzed the phenomenon of jumping rupture, but none has focused on how the process of rupture termination on the primary (nucleating) fault segment affects the ability of rupture to jump to a secondary fault segment. In the current study, I model the dynamics of a simple 2D strike-slip fault system with two parallel segments ar- ranged with either a compressional or extensional stepover. I vary the suddenness with which the initial shear stress tapers to zero on the primary section. If the initial shear stress goes to zero over a very small (100 m) distance, rupture readily jumps both compressional and extensional stepovers of 1 km. If the initial shear stress tapers to zero over 1 km, rupture can jump the compressional stepover, but not the exten- sional stepover. If the initial shear stress tapers to zero over 2.5 km, rupture cannot jump either the compressional or the extensional stepover. The results illustrate the importance of the slip gradient (and the resultant static stress field) and the accelera- tion of the rupture front (and the resultant generation of stopping phases) in determin- ing the probability of jumping rupture.