Ratios of the peak amplitudes of simulated short-period Wood-Anderson and long-period Wood-Anderson seismograms are compared for the mainshock and the two largest aftershocks. 152 4.13 Directional analysis at GSC, ISA and PFO (Uniform slip) 155 4.14 Comparison of SH displacement spectra for the main shock and two.
Introduction to Broadband Seismology
General Introduction
To illustrate this point, Figure 1.1 compares the tangential component displacement record from the 1990 Upland, California earthquake (Mw = 5.5), recorded at Pasadena (distance 44 km), with simulated Wood-Anderson (WASP) records. The first WASP record (Figure 1.1a) has a clip level set to 12.7 cm, which is half the width of the recording paper on a heliocorder.
Tangential Component, 1991 Baja California
The first part deals with the inversion of three-component data for 5 of these stations. However, this model does not explain the amplitudes of short-period waves (> 1 Hz).
Determination of Sourc e Parameters at Regional Distances with Three
Component Sparse Network Data
Introduction
However, smaller events are recorded with poor signal-to-noise ratios at distances greater than 30°. Although the regional waveforms are relatively more complicated, the long‐period Pnl waves are relatively stable (i.e., change slowly with distance) [ Heimberger and Engen , 1980 ] and have been shown to be quite successful in source inversions [ Wallace et al. , 1981].
Case for the Standard Southern California Velocity Model
Of greater importance, however, was the similarity of the waveforms at three of the equidistant stations, indicating that a co=on-Green function was sufficient for use in source inversion. The synthetic values were calculated to 5 Hz using a slightly modified version of the standard Southern California velocity model (SC, Table 2.1) commonly used to locate earthquakes. This model is mainly based on the travel times of quarry blasts [Hadley and Kanamori, 1977) and teleseismic surface waves [Hadley and Kanamori, 1979).
West Longitude
Inversion Method
- Effects of Hypocenter Mislocation
- December 3, 1991 Baja Event
Examination of the parameter space (Figure 2.6a) reveals that an erroneous distance shift produced two globals. The addition of the second station significantly improves the parameter space, in that the local minima observed in Figure 2.6a have disappeared.
Discussion and Conclusions
The location of this event determined by CICESE places it about 20 km north of the San Miguel fault. The mechanism we obtained for the 1991 earthquake is consistent with the surface rupture observed after the 1956 event [ Shor and Roberts , 1958 ]. The errors are the standard deviation of the mean of the solutions listed in Table 2.2.
There is a fairly large uncertainty for the rake parameter, but it is possible to discern that a strike-slip fault has been broken. Their rake is also somewhat poorly limited and matches the average value we obtained. While the signal-to-noise ratios become small at long periods for events smaller than magnitude 4.5, the ratios in the frequency band from 0.1 to 1 Hz (roughly the bandwidth of the WALP instrument) remain large for smaller events.
Upland Earthquakes
Data and Processing
In this study, the displacement seismograms (0.02 to 7.0 Hz), as well as the WASP and WALP instrument seismograms are used in the forward modeling approach. Both earthquakes occurred in the Upland, California region (Figure 3.1), and the hypocenters at the Southern California Network are within 1 km of each other, at a depth of about 8 to 9 km. Because the events occurred in nearly the same location, the differences in the waveforms are likely due to different source history, orientation, and/or location.
The most significant difference in the waveforms for the two events is in the relative amplitudes of the various S-wave phases in the tangential components. The approach taken in this study was to identify phases in waveforms through forward modeling using Generalized Ray Theory (GRT). With the exception of Figure 3.9, all synthetic seismograms presented in this paper are constructed using the focal mainshock mechanism.
Modeling Results
As in Figure 3.5, the displacement amplitude of the synthetic materials was fitted to the data by multiplying by the seismic moment scaling factor, and the same factor was used for the WASP and WALP synthetic materials. As in Figure 3.5, there is good agreement even for the maximum amplitudes of the synthetics and the data collected with the WASP and WALP instruments. Also, the tangential component of the mainshock is best modeled with a source depth of 6 km and an aftershock between 8 km and 9 km.
However, the use of the aftershock mechanism improves the fit on the radial and vertical components. The shape of the Love wave has changed severely, as have the radial and vertical synthetic components. The width of the So phases is well modeled with the used source time function.
Int roduction
Of the two possible fault planes for the 1988 main shock, based on the observed orientation, the southwest trending plane is preferred [Mori and Hartzell, 1990]. All synthetics in this article were calculated using the LORS1 rate model (Table 3.1). As noted above, the 1990 sequence occurred in roughly the same location and had a focal mechanism similar to the two largest events of the 1988 sequence.
The focal depth of the 1990 mainshock determined from SCSN first motions was 5.2 km [ Hauksson and Jones , 1991 ], which is consistent with the first order depth estimate obtained above by simply comparing waveforms. This contrasts with the displacement data just discussed, suggesting that the short-period energy originates closer to the location of the 1988 aftershock. Comparison of displacement data with displacement synthetics calculated with source parameters of obtained from the inversion of long-period waves shows that the displacement registers,.
IVV'
Distributed Fault Models
All distributed fracture models shown here use the orientation of the fracture face solution and the moment obtained from the inversion of the long-period waves in which the functions of the colon source Green were used. We use the locations of aftershocks (Figures 3.14ab) to obtain the initial dimension of the distributed rupture and the location of the hypocenter. Profile BB' (Figure 2.14a) shows the orientation of the finite fracture relative to the aftershock zone.
The hypocenter is centered laterally and is located at the top of the fault, at a depth of 6 km. The center of the bump is located 0.875 km southwest of the hypocenter at a depth of 8.5 km. There is some extension of So in both the displacement and the WASP synthetics, and this model likely represents an upper bound on the bump size.
D iscussion and Conclusions
Additionally, Hauksson and Jones [1991J] point out that the aftershock zone lies southeast of the inferred Cucamonga fault trace and projects to the location of the San Jose fault. However, it should be noted that the location of the Cucamonga fault in this area is found and hidden by alluvial deposits. The San Jose fault to the east of the epicenter (Figure 3.14a) is a possible candidate as the causative structure.
It is interesting to compare the depth distribution of aftershocks (Figure 3.14b) with the location of the asperities and the source area determined by reversing the long-period waves. Considering all the events (Figure 6), it appears that there is a relatively small hole (1 km2) at 10 km depth, slightly southwest of the epicenter. It is possible that the observed fracture complexity at this depth inhibited the propagation of the main shock's fracture front.
The 1991 Sierra Madre Earthquake
Inversion of Long- Period Body Waves
- Inversion Results
- Introduction
The variation in travel time Pn is less than 2% of the average travel time of 26.2 seconds. Since the PAS was only 21 km from the epicenter, the attenuation is negligible and the direct S wave width is a good approximation to the source time function. As noted earlier, many of the shorter period phases are evident in the data, but have small discrepancies with the travel time.
There are also some discrepancies in the relative amplitudes of some body wave phases compared to surface waves. Few studies have been conducted to date to determine the source parameters of the main shock. Profile B-B' along strike shows a curious lack of aftershock seismicity west-southwest of the hypocenter.
Short to Long Period Amplitude Ratios
For example, Figure 4.9 shows the decomposition of the wave field into sets of Love, ascending and descending rays. These models have uniform slip in that each of the point sources has equal seismic moment. This model has a hypocenter in the lower NE corner of the fault (Figure 4.11) with the rupture continuing updip to the southwest, towards station PAS.
This arrival is apparently not evident in the displacement waveforms, but contributes to the observed duration of the displacement pulse. After3 and After4 have more complex velocity waveforms that contribute to the duration of the displacement pulse. In any case, as Figure 4.12 shows, the S-wave shape of the mainshock recorded at PAS is probably dominated by the directionality rather than the slip distribution.
VIM I
- Discussion and Conclusions
A comparable observation was made for the 1990 Upland earthquake, where a relatively small patch of large slip controlled the shape and amplitude of the shorter periodic waveforms recorded by PAS. Furthermore, the rise times of each of the far-field pulses were significantly smaller than necessary to explain the near-field motions. Imprecise knowledge of Q along the paths casts doubt on individual measurements, but since all the events in that study have similar travel paths, the differences between the events are likely to be meaningful.
Seismicity and tectonics of the Cucamonga fault and eastern San Gabriel Mountains, San Bernardino County. Evidence of Extensional Basin and Range Tectonics in the Sierra Nevada: The Durwood Meadows Swarm, Tulare County, California. Inversion of Regional Pnl and Surface Wave Data for Borah Peak Aftershock Source Parameters.
Comparison of Several Q Models
Standard Southern California Model Synthetics
