PROGRESS REPORT
December 1997

Crustal Setting of the Northridge Earthquake:
Analysis of the LARSE Malibu-Northridge-West Mojave Transect

PIs:

David Okaya and Tom Henyey, USC

In collaboration with the LARSE working group.

Objectives: (1) To process the Northridge offshore MCS seismic profile;
(2) To analyze the LARSE Northridge onshore-offshore transect in order
to obtain a crustal-scale velocity structure surrounding the Northridge
epicentral region.
(3) To collaborate with the USGS in their effort to obtain a crustal-scale velocity
structure along the onshore-offshore portion of the LARSE San Gabriel
transect.

The Los Angeles Region Seismic Experiment (LAPSE) is a major project within SCEC. This multi-faceted study is designed to obtain crustal-scale geometrical and velocity structural profiles along transects across the LA basin and surrounding region (Figure 1). This study represents the first high-resolution regional seismic study of the structural framework of the LA region and will contribute to the construction of the SCEC master model. USC was an active participant in the field acquisition and subsequent data reduction and archive.

Data Analysis

Seismic data processing of MCS profile Line 02 was completed during 1997. The profile in reduced form is shown in Figure 2. The profile was initially processed to 6 sec after which 12 sec stacks were produced. Little reflective energy is observed below 12 see; hence the 6 sec stack is available for distribution. Basins in the Santa Monica basin and in the floors on the sides of Santa Catalina and San Clemente islands show internal structure with strong basement contacts. Little additional reflective energy is actually observed below approx 2-3 sec in the basement other than an intriguing low-angle basement event south of San Clemente Island.

Analysis of the Northridge onshore-offshore data proceeded in 1997. With the arrival of post-doe Nicola Godfrey, USC offered to assist the USGS in the analysis of onshore-offshore data from Transect I (San Gabriel). While initial progress was made on the Northridge transect using the Zelt forward/inversion analysis approach (Zelt and Smith, 1992), with the guidance of post-doe Godfrey we switched to the Hole 2D/3D inversion method (Hole, 1992). In order to use this method on the two LARSE transects as well as for shallow industry profiles donated to SCEC, we spent a couple of months streamlining the procedures to set up and run the software associated with this method. The scripts and related utility software are available and will also be donated to John Hole for distribution with his software.

For the analysis of the Northridge transect, an initial starting model was constructed using velocity results which are already available. A shallow marine velocity structure was obtained from Uri ten Brink which was created using OBS data collected during LARSE '94 (ten Brink et al., in prep.). This structure was reparameterized and emplaced into a portion of a cross-sectional model of Transect 2. Onshore velocity information for the upper 15 km between the coastline and the Transverse Ranges was obtained from Haukkson and Haase (1997). To the extreme north, velocity information associated with the northernmost Mojave was obtained from Malin et al (1996). These velocity information for the upper 15 km were regridded and smoothed in order to be used as starting models for the velocity inversions (Figure 2).

Of the 47 common receiver gathers of Transect II, 35 have usable data; the stations located in the more heavily populated regions have little data. Pg, PmP, and Pn phases were interpreted with their travel times picked. Figure 3A illustrates the distinct Pg and PmP phases for a station in the Santa Monica Mts. Figure 3B is from a station in the central west Mojave and has prominent Pg, Pn, PmP phases. An additional phase is present behind Pn at postcritical distances which has
an unusual (slow) moveout. This phase resembles a Pg; it is observed on a number of stations within and north of the San Gabriel Mts.

Figure 4 illustrates the assembled starting model for the upper crust with three different starting velocity gradients for the lower crust. Examples of velocity structure using only the Pg and Pn phases from the onshore-offshore data are also illustrated. These figures indicate that the onshore-offshore Pg/Pn information is insufficient to characterize the lower crust; the PmP reflections are critical to resolving the geometry of Moho. The OBS first arrivals need to also be included to freeze the offshore upper crust while the OBS PmP data is needed to futher constrain Moho under the Borderlands. A critical set of data which is missing is land sources recorded into land stations - this would characterize the onshore upper crust; this data will be collected as part of LARSE II during the upcoming year. PmP reflection information has not yet but will be incorporated into the velocity analysis using compatible inversion methods (Hole and Zelt, 1995).

Transect I (San Gabriel) requires the merging of three diverse data sets into the velocity analysis. OBS data between Santa Catalina Island and Seal Beach were analyzed by ten Brink (in prep) with traveltime picks made available. The land explosion travel time picks were made available by the USGS. Onshore-offshore picks were made at USC. The merging of these three data sets required making compatible the different local coordinate frames used by each group into one overall consistent system. This integration has allowed us to begin inversion iterations using the Hole approach. The Northridge and San Gabriel transect analyses are in progress in collaboration with the LARSE working team.

Figures

Fig 1. Location map of LARSE '94 experiments.

Fig 2. MCS profile LAPSE-02.

Fig 3. Onshore - offshore data. (A) Santa Monica Mts showing excellent Pg and PmP phases. (B) central west Mojave showing Pg, PmP, Pn, and an unusual (6.0 km/s) event seen only at post-critical offsets.

Fig 4. Examples of velocity inversions using different lower crustal starting velocities. The upper crustal velocities are well defined using results from ten Brink et al. (in prep), Haukkson and Haase (1997), and Malin et al. (1995). Only the Pg and Pn phases are included here and so the lowermost crust is not well defined. Incorporation of the Moho reflection phases is required before the velocity structures may be used for interpretation.

References

Haukkson, E., and Haase, J., 1997, Three-dimensional Vp and Vp/Vs velocity models of the LosAngeles basin and central Transverse Ranges, California, JGR, 102, 5423-5454.

Hole, J., 1992, Non-linear high-resolution three-dimensional seismic travel-time tomography,JGR, 97, 6553-6562.

Hole, J., and Zelt, B., 1995, Three-dimensional finite-difference reflection traveltimes, Geophys.J. Int., 121, 427-434.

Malin, P., Goodman, E., Henyey, T., Li, Y.G., Okaya, D., and Saleeby, J., 1995, Significanceof seismic reflections beneath a tilted exposure of deep continental crust, Tehachapi Mountains,California, JGR, 100, 2069-2087.

ten Brink, U., Zhang, J., Brocher, T., Okaya, D., Klitgord, K., and Fuis, G., The innerCalifornia Borderland as a possible metamorphic core complex: seismic evidence, in prep.

Zelt, C., and Smith, R., 1992, Seismic traveltime inversion for 2-D crustal velocity structure,Geophys. J. Int., 108, 16-34.