The Source Mechanism of the August 7 , 1966 El Golfo Earthquake

The El Golfo earthquake of August 7, 1966 (mb = 6.3, Ms = 6.3) occurred near the mouth of the Colorado River at the northern end of the Gulf of California. Synthetic seismograms for this event were computed for both the body waves and the surface waves to determine the source parameters of the earthquake. The body-wave model indicated the source was a right lateral, strike-slip source with a depth of 10 km and a far-field time function 4 sec in duration. The bodywave moment was computed to be 5.0 x 102s dyne-cm. The surface-wave radiation pattern was found to be consistent with that of the body waves with a surface-wave moment of 6.5 x 102s dyne-cm. The agreement of the two different moments indicates that the earthquake had a simple source about 4 sec long. A comparison of this earthquake source with the Borrego Mountain and Truckee events demonstrates that all three of these earthquakes behaved as high stressdrop events. El Golfo was shown to be different from the low stress-drop, plateboundary events which were located on the Gibbs fracture zone in 1967 and 1974. INTRODUCTION Several studies have been reported in the literature in which long period bodywave form data have been used to model a seismic source (Mikumo, 1971a, b; Burdick and Mellman, 1976; Langston and Butler, 1976; Langston, 1976). Likewise, a number have been based on sub:face-wave data (Ben-Menahem and Toksoz, 1962, 1963a, b; Kanamori, 1970a, b). Recently Kanamori and Stewart (1976, 1978) have used both types of information to constrain the source models of several earthquakes. In order to fit both the body waves and the relatively longer period surface waves, it is necessary to find a model 'which accurately represents the behavior of the source over a long time span. While not unique, it is probable that a model which achieves this correctly accounts for the dominant equivalent elastic processes occurring at the source. A relatively simple source model can account for both types of data from some moderate sized strike-slip events such as those along the Gibbs fracture zone (Kanamori and Stewart, 1976) or the Truckee earthquake (Burdick, 1977). For others such as the Borrego Mountain earthquake, a simple model will not simultaneously account for the bodyand surface-wave moments (Burdick, 1977). The purpose of this paper is to present the results of a composite surface-wave bodywave study of the August 7, 1966 strike-slip E1 Golfo earthquake in Sonora, Mexico and to compare its source to those found for the Truckee, Borrego Mountain, and Gibbs fracture zone events. The E1 Golfo earthquake is a particularly interesting event to study because of its unique location at the northern end of the Gulf of California. It occurred in a region where the crust is relatively thin, the depth to Moho being only about 20 km (Phillips, 1964). The location of the E1 Golfo event makes it ideal for a study of its seismic radiation because the available information on the seismic crustal structure near the source (Phillips, 1964) eliminates some of the uncertainties in the bodywave modeling. The near-source structure can be assumed to be known and not changed in the body-wave modeling process. 1281 1282 JOHN E. EBEL, L. J. BURDICK, AND GORDON S. STEWART The origin time and epicenter of the E1 Golfo earthquake were listed by Sykes (1970) as: 17h36m22.8s UT and 31.72°N, 114.42°W, respectively, with mb = 6.3 and M s = 6.3. The depth was unconstrained. The epicenter was located on the Colorado River delta near the village of E1 Golfo de Santa Clara in northern Mexico (Figure 1). It occurred on or near a right-lateral strike-slip fault, here called the San Jacinto fault, which runs from Cerro Prieto, an active volcano, into the Gulf of California. A detailed description of this fault is provided by Merriam (1965). Unfortunately, there are no published reports on the effects of this tremor in the epicentral area, although there was a report of ground breaking along this fault in the Colorado delta area (J. Brune, personal communication, 1976). There was a large (ML = 7.1) earthquake in this area in 1934 which had sizable surface displacement (Biehler et al. , 1964). FIG. 1. Location of the E1 Golfo earthquake on the San Jacinto fault at the mouth of the Colorado River. The interpretation of tectonics is from Elders et al. (1972). TECTONIC SETTING The E1 Golfo earthquake occurred in a region where there is a transition from oceanic to continental tectonic features. The Gulf of California is being opened by a series of spreading centers and transform faults which are the northern extension of the East Pacific rise spreading system. In the southern Gulf there is a well developed spreading system, but in the north the details of spreading are unclear. Lomnitz et al. (1970) used details of seismicity in the northern Gulf to argue for a series of spreading centers and transform faults running from the Gulf into the Imperial Valley in California. They viewed the San Jacinto fault as a transform fault from the Gulf north to Cerro Prieto. Thatcher and Brune (197I) studied earthquake swarms in the Wagner and Delfin basins in the northern Gulf and concluded that the swarms are related to spreading in these basins. EL GOLFO EARTHQUAKE OF AUGUST 7, 1966 1283 Henyey and Bischoff (1973) argued from seismic reflection profiles in the northern Gulf that although there is much faulting, the evidence for a spreading centertransform fault system is weak. There are no magnetic stripes anywhere in the Gulf (Larson et aL, 1972), although the high sedimentation rate may be obscuring the magnetic lineations. Crustal thickness and seismic velocities are not known in the epicentral region but can be extrapolated from studies in the northern Gulf of California by Phillips (1964) and in the Imperial Valley by Kovach et al. (1962). These studies show a crust which is thickest in the Colorado delta region and which thins both north into the Imperial Valley and south into the Gulf of California. The crustal thickness is about 20 km.