The following list of recently published papers are based on research sponsored by SCEC. These papers are NOT available from SCEC. Most of the journals containing these papers are available at university libraries, and authors may also have reprints of their papers available by request.

If you are looking for technical publications available for puchase, visit our products and publications page.

Other SCEC research papers are listed in an online database.

List of publications announced 8/3/2001

List of publications announced 9/20/2001

Recent Research as of 12/18/2001:

(SCEC Contribution numbers are in bold)

521. Lewis, J.L., S.M. Day, H. Magistrale, R. Castro, L. Astiz, C. Rebollar, J. Eakins, F.L. Vernon, and J.N. Brune, Crustal thickness of the Peninsular Ranges and Gulf Extensional Province in the Californias, Journal of Geophysical Research, 106, pp. 13599-13611, 2001.

Previous crustal thickness studies in southern California, U.S.A., indicate a steeply dipping Moho beneath the topographically high eastern Peninsular Ranges. We continue these related studies to the south in northern Baja California, México, using an 11-station array that crosses the Baja California peninsula and Gulf of California at ~31ºN. Moho Ps conversions, identified in teleseismic receiver functions, are used to estimate the crustal thicknesses beneath these stations. Depth to the Moho is ~33 km near the Pacific coast of Baja California and increases gradually toward the east, reaching a maximum depth of ~43 km beneath the western part of the Peninsular Ranges batholith. The crust then thins rapidly under the topographically high eastern Peninsular Ranges and across the Main Gulf Escarpment. Crustal thickness is ~15-18 km within and on the margins of the Gulf of California. The Moho shallowing beneath the eastern Peninsular Ranges represents an average apparent westward dip of ~25º. This range of Moho depths within the Peninsula Ranges, as well as the sharp ~east-west gradient in depth in the eastern part of the range, are in agreement with the earlier observations from north of the international border. The Moho depth variations do not correlate with topography of the eastern batholith. These findings suggest that a steeply dipping Moho is a regional feature beneath the eastern Peninsular Ranges, and that a local Airy crustal root does not support the highest elevations. We suggest that Moho shallowing under the eastern Peninsular Ranges reflects extensional deformation of the lower crust in response to adjacent rifting of the Gulf Extensional Province that commenced in the late Cenozoic. Support of the eastern Peninsular Ranges topography may be achieved through a combination of flexural support and lateral density variations in the crust and/or upper mantle.

530. Olsen, K.B., Three-dimensional ground motion simulations for large earthquakes on the San Andreas fault with dynamic and observational constraints, Journal of Computational Acoustics, 9, no. 3, pp. 1203-1215, 2001.

I have simulated 0-0.5 Hz viscoelastic ground motion in Los Angeles from M 7.5 earthquakes on the San Andreas fault using a fourth-order staggered-grid finite-difference method. Two scenarios are considered: (a) a southeast propagating and (b) a northwest propagating rupture along a 170-km long stretch of
the fault near Los Angeles in a 3D velocity model.

The scenarios use variable slip and rise time distributions inferred from the kinematic inversion results for the 1992 M 7.3 Landers, California, earthquake.

The rough static slip distribution used in this study, unlike that modeled in a recent study, is in agreement with constraints provided by rupture dynamics. I find peak ground velocities for (a) and (b) of 49 cm/s and 67 cm/s, respectively, near the fault.

The near-fault peak motions for scenario (a) are smaller compared to previous estimates from 3D modeling for both rough and smooth faults. The lower near-fault peak motions are in closer agreements with constraints from precarious rocks located near the fault.

Peak velocities in Los Angeles are about 30% larger for (b) (45 cm/s) compared to those for (a) (35 cm/s).

541. Hardebeck, J. L. and Hauksson, E., The crustal stress field in southern California and its implications for fault mechanics, Journal of Geophysical Research, 106, pp. 21859-21882, 2001.

We present a new, high spatial resolution image of stress orientation in southern California based on the inversion of earthquake focal mechanisms. We use this image to study the mechanics of faulting in the plate boundary region. The stress field contains significant spatial heterogeneity, which in some cases appears to be a result of the complexity of faulting, and in other cases, a cause. Temporal changes in the stress field are also observed, primarily related to major earthquakes. The observed 15 degree rotation of the stress axes due to the 1992 M7.3 Landers mainshock implies that the deviatoric stress magnitude in the crust is low, on the order of 10 MPa. This suggests that active faults in southern California are weak. The maximum principal stress axis near the San Andreas Fault is often at 50 degrees to the fault strike, indicating that the shear stress on the fault is comparable to the deviatoric stress. The San Andreas in southern California may therefore be a weak fault in a low-strength crust.

554. Peyrat, S., K.B. Olsen, and R. Madariaga, Dynamic modeling of the 1992 Landers earthquake, Journal of Geophysical Research, 106, B11, pp. 26467-26482, 2001.

We have modeled dynamic rupture propagation for the 1992 Landers earthquake using a fourth-order finite-difference method (Madariaga et al., 1998). In this method, the dynamic rupture grows spontaneously under the simultaneous control of the initial load and rupture resistance by friction, modeled using a simple slip weakening law. We used a simplified Landers fault model where the three segments were combined into one vertical, planar fault. By trial-and-error we modified the initial stress field (Olsen et al., 1997) until the dynamic rupture generated a rupture history and final slip distribution that approximately fit those determined by the kinematic inversions by (Wald and Heaton, 1994; Cohee and Beroza, 1994; and Cotton and Campillo, 1995). We find that rupture propagation is extremely sensitive to small changes in the distribution of prestress and that energy release rate controls the average rupture speed. Clearly, dynamic models have many more degrees of freedom than kinematic inversions where the rupture velocity is typically heavily constrained. Our results suggest that, to some extent, the slip distribution and rupture velocity obtained by kinematic inversions reflect the initial stress distribution and rupture resistance, respectively. In order to validate our model we generated synthetic 0.5 Hz accelerograms from our dynamic simulation with the Green's functions propagation method (Bouchon, 1981; Kennett and Kerry, 1979). This method enables us to propagate the radiation generated by the dynamic rupture to distances greater than those feasible using the finite-difference method. We find a satisfactory fit between our synthetics and the strong motion data.

556. Ben-Zion, Y., Dynamic rupture in recent models of earthquake faults, J. Mech. Phys. Solids, 49, pp. 2209-2244, 2001.

We discuss several problems of dynamic rupture relevant to mechanics of earthquake faults, material sciences, and physics of spatially extended dissipative systems. The problems include dynamic rupture along an interface separating different elastic solids, dynamic rupture on a planar surface governed by strongly velocity-weakening friction, and elastodynamic calculations of long deformation history on a smooth fault in an elastic continuum. These separate problems share a number of methodological and conceptual issues that form recurring themes in the paper. An important methodological issue for computational schemes is dependency of numerical results on the used grid size. This arises inevitably in computer simulations when the assumed constitutive laws do not include a length scale (e.g., of shear or extensional displacement) over which material properties evolve. Such simulations do not have a stable underlying solution, to which they may converge with sufficient grid refinement. However, they may provide rough approximations -- lacking at present a rigorous foundation -- to the behavior of systems containing elements of discreteness (associated with abrupt fluctuations) at scales relevant to observations of interest. Related important conceptual issues are connections between, or when appropriate separation of, small scale phenomena (e.g., nucleation of rupture, processes at rupture front) and large scale features of the response (e.g., overall space-time dimensions of rupture, statistics of many events). Additional recurring conceptual topics are crack vs. pulse modes of dynamic rupture, the stress under which earthquake faults slip, and the origin of spatio-temporal complexities of earthquakes. These seemingly different issues probably have one or more common origins. Dynamic rupture on an interface between different solids, strongly velocity-weakening friction on a homogeneous fault, and strong fault zone heterogeneities can all produce narrow self-healing slip pulses with low dynamic stress (and low associated frictional heat) during the active part of slip. Strong fault heterogeneities probably play the dominant role in producing the observed earthquake complexities. Improved understanding of the discussed problems will require establishing connections between discrete and continuum descriptions of mechanical failure processes, generalization of current models to realistic three-dimensional dynamic models, and high-resolution laboratory and in-situ observations over broad scales of space and time. These challenging problems provide by their subject matter and involved great difficulties important targets for multi-disciplinary research by engineers, earth scientists, and physicists.

578. Nikolaidis, R.M., Y. Bock, P.J. de Jonge, P. Shearer, D.C. Agnew, and M. Van Domselaar, Seismic wave observations with the global positioning system, Journal of Geophysical Research, 106, pp. 21897-21916, 2001.

We describe the direct measurement of ground displacement caused by the Hector Mine earthquake in southern California (Mw 7.1, October 16, 1999). We use a new method of instantaneous positioning, which estimates site coordinates from only a single epoch of Global Positioning System (GPS) data, to measure dynamic as well as static displacements at 24 stations of the Southern California Integrated GPS Network (SCIGN), with epicentral distances from 50 to 200 km. For sites outside the Los Angeles basin, the observed displacements are well predicted by an elastic half-space model with a point shear dislocation; within the sedimentary basin, we observe large displacements with amplitudes up to several cm that last as long as 3-4 minutes. Since we resolve the GPS phase ambiguities and determine site coordinates independently at each epoch, the GPS data rate is the same as the receiver sampling rate. For the SCIGN data, this is 0.033Hz (once per 30 seconds), though sample rates up to 2 Hz are possible with the SCIGN receivers. Since the GPS phase data are largely uncorrelated at 1 s, a higher sampling rate would offer improved temporal resolution of ground displacement, so that, in combination with inertial seismic data, instantaneous GPS positioning would significantly increase the observable frequency band for strong ground motions.

610. Perfettini, H., J. Schmittbuhl, J. R. Rice, and M. Cocco, Frictional response induced by time-dependent fluctuations of the normal loading, Journal of Geophysical Research, 106, no. 7, pp. 13455-13472, 2001.

We study the effect of time variable normal stress perturbations on a creeping fault which satisfies a velocity-weakening rate-and state-dependent friction law and is slipping at constant speed. We use the spring block model and include the effect of inertia. To account for the variable normal stress, we use the description introduced by Linker, 1992, which links normal stress fluctuations to changes of the state variable. We consider periodic perturbations of the normal stress in time (as caused for instance by tides) and compare the behavior for two commonly used friction laws (the "slip" and the "ageing" laws). Their mechanical response is shown to be significantly different for normal stress fluctuations. It could be used to probe these two laws during laboratory friction experiments. We show that there is a resonance phenomenon, involving strong amplification of the shear and velocity response of the interface, when the spring stiffness is modestly above its critical value (or when, at a given stiffness, the normal stress is modestly below its critical value). We show that such an amplification is also observed when periodic fluctuations of the shear loading are considered, making the resonance phenomenon a general feature of the response of a near-critical creeping surface to periodic fluctuations of the external loading. Analytical solutions are based on a linear expansion for low amplitude of normal or shear stress variations and are in very good agreement with numerical solutions. A method to get the evolution of friction in the case of an arbitrary perturbation of the normal stress is also presented. The results show that a creeping fault may be destabilized and enter a stick-slip regime owing to small normal stress oscillations. This may also account for a mechanism for the generation of "creep bursts". However, these phenomena require very specific parameter ranges, to excite the resonance, which may not be met very generally in nature. This study illustrates the importance of the normal stress fluctuations on stable sliding and suggests further friction laboratory experiments.

622. Spudich, P. and K.B. Olsen, Fault zone amplified waves as a possible seismic hazard along the Calaveras fault in central California, Geophysical Research Letters, 28, pp. 2533-2536, 2001.

The Calaveras fault is situated within a low velocity zone (LVZ) 1-2 km wide near Gilroy, California. Strong motion accelerographs G06, located in the LVZ 1.2 km from the Calaveras fault, and G07, 4 km from G06, recorded both the M 6.2 1984 Morgan Hill and the M 6.9 1989 Loma Prieta earthquakes. Comparison of the ground motions shows that a large 0.6-1.0 Hz velocity pulse observed at G06 during the Morgan Hill event may be amplified by focusing caused by the LVZ. Such amplified waves might be a mappable seismic hazard, and the zone of increased hazard can extend as much as 1.2 km from the surface trace of the fault. The discrepancy between the inferred wide LVZ on the Calaveras and the narrow LVZs inferred on other faults might be actual, making the Calaveras an unusual fault zone, or might result from systematic differences of observational technique. The observed zone of ground motion amplification is not contained within the Special Studies Zone defined at this location.

628. Zeng, Y. and C. Chen, Fault rupture process of the September 20, 1999 Chi Chi, Taiwan earthquake, Bulletin of the Seismological Society of America, 91, pp. 1088-1098, 2001.

We modeled the source rupture process of the September 20, 1999 Chi Chi, Taiwan earthquake (Mw7.6) using an integrated dataset of near-field strong ground motion and GPS observations. This large inland thrust earthquake produced more than 80 km long surface break along the Chelungpu fault. The event triggered over seven hundred strong motion instruments on the Island of Taiwan. This large ground motion dataset provides the best opportunity to date to understand earthquake source processes in a major earthquake. First, we develop a 3-D curved fault model consistent with the surface rupture data and aftershock distribution. The subsurface fault plane has an average dip of 30 degrees to the east, and measures 96 km along strike and 40 km down-dip. Next we applied a genetic algorithm to reconstruct the source rupture process. The inversion results indicate a complex slip distribution with a generally broad triangular shaped slip zone that points down-dip. Most of the displacement occurred over a 15 km depth range in the northern segment of the fault, with the maximum slip exceeding 20 m. The average slip is 3.8 m and the average rupture velocity is 2.6 km/sec. The mainshock slip distribution clearly complements the aftershock location distribution. Slip rise times range from 8 to 6 seconds over most of the primary slip area and decrease slightly as rupture propagates south to north. This relatively long rise time resulted in a low dynamic stress drop and low peak ground accelerations observed at surface stations. The rake angle rotates across the rupture plane from predominantly dip-slip in the southern segment to mostly left-lateral oblique slip in the north. We find that geometrical irregularities over the fault plane played an important role in controlling the temporal rupture behavior. At fault bends, rupture tends to decelerate, the rake angle rotates to a near dip-slip orientation and the shear stress increases as rupture tries to overcome the bending barriers. The total seismic moment from our inversion is 2.9x1027 dyne-cm, in good agreement with the Harvard and USGS moment estimation.

629. Zeng, Y., Viscoelastic stress-triggering of the 1999 Hector Mine earthquake by the 1992 Landers Earthquake, Geophysical Research Letters, 28, pp. 3007-3010, 2001.

The 1999 Mw 7.1 Hector Mine earthquake occurred only 20 km from the 1992 Mw 7.3 Landers earthquake fault. The close spacing between the two earthquakes has lead to wide speculation that the two events are related. The main question is whether the Landers earthquake triggered the Hector Mine event. The condition under which faults break is usually characterized by the Coulomb failure criterion. A stress transfer calculation by scientists from the USGS, SCEC and CDMG (2000) using an elastic half-space model has shown, however, a negative Coulomb stress change of -1.4 to -0.6 bar at the Hector Mine hypocenter due to Landers. A negative Coulomb stress change would be inconsistent with the hypothesis of triggering by static stress, suggesting that other physical mechanisms have controlled the process of the stress evolution after the Landers earthquake. In this paper, I show evidence of stress triggering of the Hector Mine earthquake by Landers, due to a process governed by viscoelastic flow in the lower crust. This visoelastic flow has produced broad-scale postseismic rebound observed by GPS and InSAR measurements (Deng et al., 1998). The result of this study is that viscoelastic flow has significantly modified the regional stress field in the Mojave Desert after the Landers earthquake. The evolving stress field, including viscoelastic flow, has gradually moved the Coulomb stress change at the Hector Mine hypocenter to a positive level. The increase in Coulomb stress exceeded 1 bar right before the Hector Mine earthquake, bringing the Hector Mine ruptures to the proximity of catastrophic failure.