Numerical modeling of earthquake dynamics and derived insight for seismic hazard relies on credible, reproducible model results. The sequences of earthquakes and aseismic slip (SEAS) initiative has set out to facilitate community code comparisons, and verify and advance the next generation of physics-based earthquake models that reproduce all phases of the seis-mic cycle. With the goal of advancing SEAS models to robustly incorporate physical and geo-metrical complexities, here we present code comparison results from two new benchmark problems: BP1-FD considers full elastodynamic effects, and BP3-QD considers dipping fault geometries. Seven and eight modeling groups participated in BP1-FD and BP3-QD, respectively, allowing us to explore these physical ingredients across multiple codes and better understand associated numerical considerations. With new comparison metrics, we find that numerical resolution and computational domain size are critical parameters to obtain matching results. Codes for BP1-FD implement different criteria for switching between quasi-static and dynamic solvers, which require tuning to obtain matching results. In BP3-QD, proper remote boundary conditions consistent with specified rigid body translation are required to obtain matching surface displacements. With these numerical and mathematical issues resolved, we obtain excellent quantitative agreements among codes in earthquake interevent times, event moments, and coseismic slip, with reasonable agreements made in peak slip rates and rupture arrival time. We find that including full inertial effects generates events with larger slip rates and rupture speeds compared to the quasi-dynamic counterpart. For BP3-QD, both dip angle and sense of motion (thrust versus normal faulting) alter ground motion on the hanging and foot walls, and influence event patterns, with some sequences exhibiting similar-size character-istic earthquakes, and others exhibiting different-size events. These findings underscore the importance of considering full elastodynamics and nonvertical dip angles in SEAS models, as both influence short-and long-term earthquake behavior and are relevant to seismic hazard.
To assess the long‐term safety of a deep repository of spent nuclear fuel, upper bound estimates of seismically induced secondary fracture shear displacements are needed. For this purpose, we analyze a model including an earthquake fault, which is surrounded by a number of smaller discontinuities representing fractures on which secondary displacements may be induced. Initial stresses are applied and a rupture is initiated at a predefined hypocenter and propagated at a specified rupture speed. During rupture we monitor shear displacements taking place on the nearby fracture planes in response to static as well as dynamic effects. As a numerical tool, we use the 3Dimensional Distinct Element Code (3DEC) because it has the capability to handle numerous discontinuities with different orientations and at different locations simultaneously. In tests performed to benchmark the capability of our method to generate and propagate seismic waves, 3DEC generates results in good agreement with results from both Stokes solution and the Compsyn code package. In a preliminary application of our method to the nuclear waste repository site at Forsmark, southern Sweden, we assume end‐glacial stress conditions and rupture on a shallow, gently dipping, highly prestressed fault with low residual strength. The rupture generates nearly complete stress drop and an Mw 5.6 event on the 12 km2 rupture area. Of the 1584 secondary fractures (150 m radius), with a wide range of orientations and locations relative to the fault, a majority move less than 5 mm. The maximum shear displacement is some tens of millimeters at 200 m fault‐fracture distance.
Using spectral amplitudes from the South Iceland Lowland (SIL) seismic network, we conduct a relative moment tensor inversion (RMTI) on aftershocks of the June 1998 M-w 5: 4 event that occurred at the Hengill triple junction, southwest Iceland. Three distinct groups of spatially clustered events are observed in the region for 25 selected events that occurred during the period from 4-5 June 1998. These clusters have previously been relocated with very high accuracy using cross-correlation techniques. We use the RMTI method to determine the focal mechanisms of these events and compare our results with the SIL network mechanisms obtained using spectral amplitudes. Most focal mechanisms obtained in this study show a predominantly right-lateral strike-slip motion, similar to those obtained by the SIL network, but more consistently in agreement with the orientations of the surface faults in the area. The spectral amplitude grouping method was used to investigate discrepancies between some of the focal mechanisms obtained using RMTI and the method used in the SIL network. This resolved apparent differences in the focal mechanism solutions for two of the studied events. Cluster alignment across the presumed fault and the individual event mechanisms agree well, suggesting the occurrence of the events along a fault plane dipping steeply towards the east. Consistency in the pressure and tension axes of the focal mechanisms suggests that the region was under northeast-southwest-oriented compression during the activity. Decomposition of the moment tensors into double-couple and isotropic components and the resulting insignificant isotropic component also suggests that the styles of failure for the analyzed events was mainly due to shearing.
We present a modeling technique for generating synthetic ground motions, aimed at earthquakes of design significance for critical structures and ground motions at distances corresponding to the engineering near field, in which real data are often missing. We use dynamic modeling based on the finite‐difference approach to simulate the rupture process within a fault, followed by kinematic modeling to generate the ground motions. The earthquake source ruptures were modeled using the 3D distinct element code (Itasca, 2013). We then used the complete synthetic program by Spudich and Xu (2002) to simulate the propagation of seismic waves and to obtain synthetic ground motions. In this work, we demonstrate the method covering the frequency ranges of engineering interests up to 25 Hz and quantify the differences in ground motion generated. We compare the synthetic ground motions for distances up to 30 km with a ground‐motion prediction equation, which synthesizes the expected ground motion and its randomness based on observations. The synthetic ground motions can be used to supplement observations in the near field for seismic hazard analysis. We demonstrate the hybrid approach to one critical site in the Fennoscandian Shield, northern Europe.
To optimize magnitude estimation for the earthquake early warning system around the Tehran region, different amplitude-and frequency-based parameters, that is, predominant period (tau(max)(p)), characteristic period (tau(c)), log-average period (tau(log)), and peak displacement (P-d) were analyzed in this article. All parameters were calculated directly from seismic records, with an epicentral distance less than 150 km, and within the initial 3 s of the P waves. The analysis of earthquakes in the 2.4 < M-L< 4.9 magnitude range verified that the result of tau(max)(p) showed a consistent trend as compared with the global observations, and provided a robust estimate of magnitude for the dataset used in this research. In comparison with worldwide observations, the calculated P-d and tau(c) were underestimated, and there was no scaling relationship with the tau(log) parameter. When combined with the global observations from Japan, Taiwan, and Italy, the results of P-d and tc for the Tehran region produced optimized results.
Ambient seismic-noise correlation is a powerful tool for extracting the seismic core phases that propagate through the interior of the Earth. In this study, we present and refine the root-mean-square-stacking method to extract stable core phases (e.g., PcP, ScS, PcS/ScP, and PKiKP) from within the Central Alborz region, Iran, using empirical Green's functions. Our studies on the extracted core phases using empirical Green's functions indicated that the ambient seismic noise method is independent from global seismicity (M >= 5.5). We also show that, by dividing ambient seismic records into shorter (i.e., 2700 s) and overlapping (70%) time windows, before the cross-correlation procedure, we can improve the quality and stability of the empirical Green's functions generated. Consequently, 73 days (equivalent to 22% of the total time period for the dataset) of nonconsecutive ambient seismic-noise time windows have been used to retrieve core-phase empirical Green's functions in a 5-10 s period band.
A relative moment tensor inversion technique is used to retrieve the focal mechanisms of teleseismic earthquakes. The observed data consists of amplitude spectra of the direct P phase on vertical and the direct S phase on rotated horizontal components. The effect of propagation paths is minimized using relative amplitude spectra of close-lying earthquakes recorded by common stations. The inversion is carried out for six components of the moment tensors. The focal mechanisms are determined using a linear weighted least-squares approach (signed spectral moment). The method is applied to seven earthquakes that occurred close to the 26 December 2003 M-w 6.6 Iran-Bam earthquake and were recorded at regional-teleseismic distances with good azimuthal coverage. Most focal mechanisms obtained in this study show a predominantly right-lateral strike-slip component, similar to those obtained by other independent estimates. The application of the relative moment tensor inversion using spectral amplitudes combined with polarity data proves to be flexible and effective in estimating focal mechanisms of teleseismic data and the method is suitable for routine implementation in seismic networks.
Here, all three components of the seismic signal are applied for use with the amplitude source location (ASL) method to investigate if using all three components yield more accurate results than using just the vertical component. Eight active source events along a debris flow channel on Te Maari Volcano, New Zealand, are used as known source locations to conduct the test. Both coda-wave normalization (CWN) and horizontal-to-vertical (HN) ratio methods are used to calculate amplification factors for station corrections. Average location errors for all the active seismic sources varied between 0.47 km for the vertical component and 0.51 km for three components while using the CWN method, and 0.92 km (vertical) and 0.83 km (three component) using the H/V method. We also conduct statistical analysis through an F-test by calculating root mean square errors (RMSEs) to determine if the results were statistically different. The RMSE analysis for the active source events shows location results for event 1 and 7 producing errors of 2.18 +/- 1.33 and 2.37 +/- 1.29 km for the vertical-component results, and 2.06 +/- 1.16 and 2.33 +/- 1.24 km for the three-component results. The F-test indicates that active source events higher up the debris flow channel (centrally located relative to the network) are statistically the same, whereas events lower down the channel (away from the center of the network) are statistically different. Results show that using all three components with the ASL method may not necessarily yield more accurate locations, but nevertheless may average the components to eliminate the extreme error values or amplify the signals, producing more precise results.