Publications

Export 4 results:
Sort by: Author Title Type [ Year  (Asc)]
1994
Scott, JS, Masters TG, Vernon FL.  1994.  3-D Velocity Structure of the San-Jacinto Fault Zone near Anza, California .1. P-Waves. Geophysical Journal International. 119:611-626.   10.1111/j.1365-246X.1994.tb00145.x   AbstractWebsite

Seismic arrival times from microearthquakes (M(L) < 4) On the San Jacinto fault near Anza, California, are used to find spatial variations in the seismic velocity that are related to the crustal structure of the fault zone. Preliminary modelling of the 1-D P-wave velocity structure of the upper 25 km of crust reveals that most of the variation in velocity is lateral rather than depth dependent. The traveltime anomalies due to lateral structure can be partially compensated for by applying station corrections, however the variance of the traveltime residuals is still 2.25 times larger than the variance of the picking error. The spatially correlated residuals show that this variance is due to localized velocity anomalies and that the data require further modelling using a 3-D velocity structure. Because the 3-D inverse problem is non-unique, smoothness constraints are applied to find the model that has the minimum structure required to fit the data to the picking error, where a smooth model is defined such that the gradient of the velocity perturbation from the original 1-D model is small. With small non-zero station corrections, a 3-D velocity model can be found that fits the data well. The structure is well resolved from 3 to 9 km depth where lateral perturbations of up to 7 per cent are determined independently of the trade-off between station corrections and poorly resolved near surface structure. The model shows a horizontal gradient with overall faster velocities in the north-east side of the fault zone. At 3-6 km depth, the signature of the fault zone is evident in the lower velocities beneath the surface trace of the fault. However, at 9 km depth, higher seismic velocities are found extending into the fault zone from the north-east block. This higher velocity region occurs where there is a distinct lack of seismicity on the fault. There is also a localized feature in the south-west of the modelled region that is more than 10 km from the main trace of the fault with velocities 3 per cent slower than average.

1997
Li, YG, Aki K, Vernon FL.  1997.  San Jacinto fault zone guided waves: A discrimination for recently active fault strands near Anza, California. Journal of Geophysical Research-Solid Earth. 102:11689-11701.   10.1029/97jb01050   AbstractWebsite

We deployed three 350-m-long eight-element linear seismic arrays in the San Jacinto Fault Zone (SJFZ) near Anza, California, to record microearthquakes starting in August through December 1995. Two arrays were deployed 18 km northwest of Anza, across the Casa Loma fault (CLF) and the Hot Springs fault (HSF) strands of the SJFZ. The third array was deployed across the San Jacinto fault (SJF) in the Anza slip gap. We observed fault zone guided waves characterized by low-frequency, large amplitudes following S waves at the CLF array and the SJF array for earthquakes occurring within the fault zone. However, we did not observe guided waves at the HSF array for any events. The amplitude spectra of these guided waves showed peaks at 4 Hz at the CLF and 6 Hz at the SJF, which decreased sharply with the distance from the fault trace. In contrast, no spectral peaks at frequency lower than 6 Hz were registered at the HSF array. We used a finite difference method to simulate these guided modes as S waves trapped in a low-velocity waveguide sandwiched between high-velocity wall rocks. The guided mode data are adequately fit by a waveguide on the CLF with the average width of 120 m and S velocity of 2.5 km/s, about 25% reduced from the S velocity of the surrounding rock; this waveguide becomes 40 to 60 m wide with the S velocity of 2.8 km/s in the Anza slip gap. On the other hand, there is not a continuous waveguide on the HSF at depth. Locations of the events with guided modes suggest that the fault plane waveguide extends along the CLF between the towns of San Jacinto and Anza, dipping northeastward at 75 degrees-80 degrees to a depth of about 18 km; it becomes nearly vertical in the Anza gap. We speculate that the existence of a continuous low-velocity waveguide on the CLF can be caused by the rupture of the magnitude 6.9 earthquake on April 21, 1918,occurring near the towns of San Jacinto and Hemet. Further, the lack of a clear waveguide on the HSF suggests that it was not ruptured in this event.

2001
Li, YG, Vernon FL.  2001.  Characterization of the San Jacinto fault zone near Anza, California, by fault zone trapped waves. Journal of Geophysical Research-Solid Earth. 106:30671-30688.   10.1029/2000jb000107   AbstractWebsite

We installed three 350-m-long seismic arrays, each array consisting of 12 three-component stations, across the Coyote Creek fault (CCF), Clark Valley fault (CVF), and Buck Ridge fault (BRF) of the San Jacinto fault zone (SJFZ) near Anza, California, to record fault zone trapped waves from microearthquakes. We observed trapped waves with relatively large amplitudes and long duration at stations close to the fault traces for earthquakes occurring within the fault zone. The coda-normalized amplitude spectra of trapped waves showed peaks at 4-7 Hz, which decreased sharply with the distance from the fault trace. Observations and three-dimensional finite difference simulations of trapped waves revealed low-velocity and low-Q waveguides on these active faults with the width of 75- 100 m in which shear velocities are reduced by 25-30% from wall rock velocities and Q values are 40-90 at depths between the surface and 18 km. The locations of earthquakes for which we observed trapped waves delineate the most seismically active fault strands of the SJFZ in a region with complicated slip planes near Anza. The low-velocity waveguides inferred from trapped waves extend 15 to 20 km in the length on these active faults and are segmented by the fault discontinuities. The waveguide on the BRF dips southwestward to connect the waveguide on the CVF, which dips northeastward. This waveguide extends at the seismogenic depth through Anza slip gap to another low-velocity waveguide on the Casa Loma fault (CLF), which has been delineated in our previous study of the SJFZ using trapped waves [Li et al., 1997]. The waveguide on the CCF in Coyote Mountain is nearly vertical and disconnected from the CLF at the south edge of Anza gap. We interpret the low-velocity waveguides on these active strands to partly result from recent prehistoric significant earthquakes on them and evaluate the future earthquake in the Anza region.

2017
Donner, S, Lin CJ, Hadziioannou C, Gebauer A, Vernon F, Agnew DC, Igel H, Schreiber U, Wassermann J.  2017.  Comparing direct observation of strain, rotation, and displacement with array estimates at Pinon Flat Observatory, California. Seismological Research Letters. 88:1107-1116.   10.1785/0220160216   AbstractWebsite

The unique instrument setting at the Pinon Flat Observatory in California is used to simultaneously measure 10 out of the 12 components, completely describing the seismic-wave field. We compare the direct measurements of rotation and strain for the 13 September 2015 M-w 6.7 Gulf of California earthquake with array-derived observations using this configuration for the first time. In general, we find a very good fit between the observations of the two measurements with cross-correlation coefficients up to 0.99. These promising results indicate that the direct and array-derived measurements of rotation and strain are consistent. For the array-based measurement, we derived a relation to estimate the frequency range within which the array-derived observations provide reliable results. This relation depends on the phase velocity of the study area and the calibration error, as well as on the size of the array.