Arben Pitarka

Simulation of Near-Fault Strong-Ground Motion Using Hybrid Green's Functions

The recently proposed hybrid Greens function method is designed to combine the advantages of both deterministic and stochastic approaches to simulating broadband ground motion when records of small events are not available. The method has the flexibility of incorporating complexities in the source, wave path, and local- site effects into strong ground motion simulations. In this article we analyze its effectiveness at simulating near-fault ground motions by comparisons with the em- pirical source time function method, empirical ground-motion-attenuation relations, and recorded near-fault ground motion. We present a simple model for introducing the effect of the radiation pattern to the stochastic Greens functions in the inter- mediate frequency range (1-3 Hz). The numerical test results of the method and the generally good agreement between simulated and recorded ground motion from the 17 January 1995 Kobe earthquake shown in this study indicate that the technique has the capability of reproducing the main characteristics of near-fault ground mo- tion.

Model for Basin Effects on Long-Period Response Spectra in Southern California

We propose a model for the effect of sedimentary basin depth on long-period response spectra. The model is based on the analysis of 3-D numerical simulations (finite element and finite difference) of long-period (2–10 s) ground motions for a suite of sixty scenario earthquakes (Mw 6.3 to Mw 7.1) within the Los Angeles basin region. We find depth to the 1.5 km/s S-wave velocity isosurface to be a suitable predictor variable, and also present alternative versions of the model based on depths to the 1.0 and 2.5 km/s isosurfaces. The resulting mean basin-depth effect is period dependent, and both smoother (as a function of period and depth) and higher in amplitude than predictions from local 1-D models. The main requirement for the use of the results in construction of attenuation relationships is determining the extent to which the basin effect, as defined and quantified in this study, is already accounted for implicitly in existing attenuation relationships, through (1) departures of the average “rock” site from our idealized reference model, and (2) correlation of basin depth with other predictor variables (such as Vs30).

Three-Dimensional Finite-Difference Waveform Modeling of Strong Motions Observed in the Sendai Basin, Japan

We perform three-dimensional (3D) finite-difference (FD) waveform modeling of strong motions in the frequency range 0.2 to 1.67 Hz observed in the Sendai basin, Japan, during the Japan Meteorological Agency magnitude ( M J) 5.0 1998 Miyagiken-Nanbu earthquake. In a previous, we estimated S -wave velocity structures above the pre-Tertiary bedrock at six sites in the Sendai basin based on array records of microtremors. To interpolate these velocity structures in space we conduct single-station microtremor measurements at a total of 61 locations and estimate the S -wave velocity structure at each site by modeling the horizontal-to-vertical spectral ratio of fundamental mode Rayleigh waves. An initial 3D model of the basin is constructed using the velocity structures estimated from both array and single-station microtremor measurements, along with other information such as surface geology. This model encompasses a region 33 km long, 30 km wide, and 19 km deep. The final model is obtained through a trial-and-error process by fitting 3D FD synthetic waveforms to the bandpass-filtered (0.2 Hz to 1.67 Hz) displacement records 12 stations for the 1998 Miyagiken-Nanbu earthquake. We compute the synthetics using a fourth-order staggered-grid 3D FD method with variable grid spacing. As the 3D model is modified, the source parameters (strike, dip, rake, and seismic moment) are estimated by a grid search method using 1D site-specific models derived from the modified 3D model. The observed waveforms are reproduced well at most stations by the final 3D basin model. This agreement suggests the validity of the final 3D basin model for theoretical strong-motion prediction of large earthquakes in the frequency range from 0.2 to 1.67 Hz. Manuscript received 7 June 2000.

On Scaling of Fracture Energy and Stress Drop in Dynamic Rupture Models: Consequences for Near‐Source Ground‐Motions

We calculate spontaneous dynamic rupture models for several well-recorded moderate to large earthquakes and analyze the scaling properties of fracture energy and stress drop. Among the set of 12 source models for 9 different earthquakes, the large events did break the surface while the moderate-size events occurred as completely buried ruptures (i.e. no surface faulting). We find that dynamic and static stress drop differ by only about 10%. Fault-averaged stress drop increases with increasing earthquake magnitude, while also fault-averaged (or maximum) fracture energy grows with magnitude. The scaling of fracture energy with the stress intensity factor appears to be sensitive to whether or not the earthquake rupture broke the surface, indicating that large earthquakes consume more fracture energy as the rupture expands and reaches the surface. This scaling of fracture energy may shed light on the recent observation that large, surface breaking earthquakes apparently generate lower near-source ground motions than buried ruptures in a certain period range of engineering interest. The derived empirical scaling relations for fracture energy may help to constrain the initial conditions for future dynamic rupture modeling, but can also be used in physics-based source characterization for near-source ground-motion calculations.

Numerical Study of Ground-Motion Differences between Buried-Rupturing and Surface-Rupturing Earthquakes

Recent ground-motion observations suggest that surface-rupturing earth- quakes generate weaker near-fault ground motion than buried earthquakes. This dif- ference is significant in the period range of 0.3-3 sec. Contributing factors to this phenomenon may include the effect of fault zone weakness at shallow depth on rup- ture dynamics and rupture directivity during earthquakes. We present results from numerical experiments of spontaneous dynamic rupture and near-source ground-motion simulations of surface rupturing and buried earth- quakes and discuss mechanisms for the observed ground-motion differences. The surface-rupturing earthquake is modeled with a shallow zone of 5 km thickness con- taining areas of negative stress drop (within the framework of the slip-weakening fric- tion model) and lower rigidity. Surface-rupturing models with this weak zone generate lower amplitude ground velocity than do models without this modification. Observed ground-motion differences between surface and buried events are qual- itatively reproduced by imposing higher stress drop in the buried earthquakes than in the surface earthquakes, combined with introducing a deeper rupture initiation for buried rupture, enhancing upward rupture-directivity effects for the latter events. In the context of our simplified model parameterization, then, the observed differences in ground motion could arise from combined effects of relative weakness of the shal- low layer of faults, the relatively larger stress drops of buried ruptures, and a tendency of near-fault sites to record strong upward directivity from buried ruptures.

Validation of a 3D Velocity Model of the Puget Sound Region Based on Modeling Ground Motion from the 28 February 2001 Nisqually Earthquake

In this study we prepared a 3D velocity model suitable for modeling long-period wave propagation in the Puget Sound region. The model is based on products of the Seismic Hazard Investigation in Puget Sound (SHIPS) and geophysical information from other studies of the region. The adequacy of the velocity model was evaluated based on analyses of goodness of fit between recorded and simulated ground-motion velocity from the M 6.8 Nisqually earthquake. The earthquake was located about 60 km south of Seattle with a hypocentral depth of 59 km. The analyses were performed in the frequency range of 0.02-0.5 Hz, using data from 40 stations. Although our model covers a wide area of the Puget Sound region, its quality is assessed in the Seattle region in which the distribution of stations that recorded the Nisqually earthquake was denser. Our 3D finite-difference ground-motion modeling suggests that the propagation of long-period waves (periods longer than 3 sec) in the Seattle basin is mostly affected by the deep basin structure. The tomographic velocity model of Parsons et al. (2001), combined with the model of depth to the basement of the Seattle basin of Blakely et al. (1999), was essential in preparing and constraining geometrical features of the proposed velocity model. Manuscript received 20 August 2003.

Ground-Motion Attenuation from the 1995 Kobe Earthquake Based on Simulations Using the Hybrid Green's Function Method

Because of the limited number of strong-motion stations in the Kobe area at the time of the 1995 Kobe earthquake, information about the characteristics of the near-fault ground-motion acceleration in bedrock is sparse. In this study we estimated the near-fault ground motion and derived characteristics of its attenuation on rock, using an hybrid broadband technique and a source model that have been validated against data. We found that at high frequencies the near-fault ground motion produced by the Kobe earthquake was of the same level as that predicted by the empirical attenuation relation for Japanese crustal earthquakes. The areas with the largest peak horizontal acceleration are located at the extremities of the fault and include most of the Kobe city.

Effects of Irregular Structure of the Mississippi Embayment on Ground-Motion Amplification

The objective of this study is to evaluate effects of the presently available 3D velocity model of the Mississippi embayment structure on the amplification of seismic waves by using simulated finite-difference seismograms. Effects of both 2D and 3D embayment basement structures were considered. The 3D model included information of the near-surface velocities that were derived from the existing 1D velocity models. The 2D crustal model was taken from Catchings (1999), which extended from Saint Louis, Missouri, to Memphis, Tennessee. Finite-difference seismograms were simulated for point sources embedded at both ends of the 2D structure. These seismograms were examined to distinguish features like peak amplitude amplifications and duration of seismograms when the seismic waves propagated from the Mississippi Embayment toward the Illinois basin and vice versa. To establish a working 3D structure model of the embayment, we compiled geologic information of the region on material properties of the shallow structure and used the 3D model developed at the Center of Earthquake Research Institute (ceri), Memphis, as the starting model. The 3D model was used to generate finite-difference seismograms along several profiles for a M w 7.2 scenario earthquake occurring on the New Madrid fault zone. An equivalent 1D model, which included the basin materials, was also used to compare the 3D versus equivalent 1D ground motions simulated using the finite-difference method. To establish the amplification factors due to the surface sediments in the 3D model, finite-difference seismograms were also computed for a 1D hard-rock reference model. These 1D and 3D responses of the Mississippi embayment were used for estimating ground-motion amplification at sites where the depth to the basement is deeper than 500 m. Our investigation suggests that the deep structure of the Mississippi Embayment has little impact on long-period ground- motion amplitudes ( T ≥ 2 sec) for large earthquakes that rupture in the central part of the basin. This supports the hypothesis that for engineering purposes ground motions simulated based on the equivalent 1D crustal model are adequate for representing ground motions from future large earthquakes ( M w > 7) occurring in the New Madrid seismic zone.

Running SW4 On New Commodity Technology Systems (CTS-1) Platform

We have recently been running earthquake ground motion simulations with SW4 on the new capacity computing systems, called the Commodity Technology Systems - 1 (CTS-1) at Lawrence Livermore National Laboratory (LLNL). SW4 is a fourth order time domain finite difference code developed by LLNL and distributed by the Computational Infrastructure for Geodynamics (CIG). SW4 simulates seismic wave propagation in complex three-dimensional Earth models including anelasticity and surface topography. We are modeling near-fault earthquake strong ground motions for the purposes of evaluating the response of engineered structures, such as nuclear power plants and other critical infrastructure. Engineering analysis of structures requires the inclusion of high frequencies which can cause damage, but are often difficult to include in simulations because of the need for large memory to model fine grid spacing on large domains.

Three-Dimensional Finite Difference Simulation of Ground Motions from the August 24, 2014 South Napa Earthquake

We performed three-dimensional (3D) anelastic ground motion simulations of the South Napa earthquake to investigate the performance of different finite rupture models and the effects of 3D structure on the observed wavefield. We considered rupture models reported by Dreger et al. (2015), Ji et al., (2015), Wei et al. (2015) and Melgar et al. (2015). We used the SW4 anelastic finite difference code developed at Lawrence Livermore National Laboratory (Petersson and Sjogreen, 2013) and distributed by the Computational Infrastructure for Geodynamics. This code can compute the seismic response for fully 3D sub-surface models, including surface topography and linear anelasticity. We use the 3D geologic/seismic model of the San Francisco Bay Area developed by the United States Geological Survey (Aagaard et al., 2008, 2010). Evaluation of earlier versions of this model indicated that the structure can reproduce main features of observed waveforms from moderate earthquakes (Rodgers et al., 2008; Kim et al., 2010). Simulations were performed for a domain covering local distances (< 25 km) and resolution providing simulated ground motions valid to 1 Hz.