MAGNITUDE 5.9 North Palm Springs Earthquake -- JULY 8, 1986

Riverside County, California

MAGNITUDE 5.9

By

GLENN BORCHARDT, Soil Mineralogist

MICHAEL W. MANSON, Geologist

Division of Mines and Geology

Pleasant Hill, California

The authors of this article were in southern California investigating the Camp Rock and Emerson faults north of Palm Springs for the Alquist-Priolo Special Studies Zones Program when the Palm Springs earthquake occurred in the early morning on July 8, 1986. They describe the event as they felt it and the lifeline damage they observed ................. editor.

INTRODUCTION

At 2:21 a.m. on July 8, 1986, we were rudely awakened in Yucca Valley by an earthquake centered on the San Andreas fault 24 km to the southwest (Figure 1). We had established temporary quarters in the area to evaluate the Camp Rock and Emerson faults to the north. The shaking of the walls of the motel was followed by an eerie silence and total darkness. The electrical power and the telephones were reconnected at about 3:00 a.m., but were disrupted again at 4:07 a.m. as we attempted to get news of the quake via television and telephone. We had no way of knowing at the time that this hour's worth of power was brought to us by a bank of 19-year-old, relatively earthquake-resistant transformers and insulators. The more-modern equipment at the Devers substation (Southern California Edison Company), which serves Yucca Valley, had failed as a result of the earthquake. The older equipment was used to provide limited power shortly after the quake, but this system also went off-line when a 220,000 volt circuit breaker opened and could not be reclosed.

FIGURE 1

Figure 1. Regional fault map (modified from Jennings, 1977) showing the epicenter of the North Palm Springs July 8, 1986 earthquake. Locations of strong-motion instruments discussed in text are also shown.

Power disruption is one of the first reliable signs that an earthquake is large enough or close enough to be really significant. In this case we were among 100,000 people who were left without power for several hours. The Devers substation sustained half the economic damage involved in the quake---over $3.5 million (Photos 1-4).

PHOTO 1

Photo 1. Devers substation (Southern California Edison Company). Insulators built in 1967 were not damaged.

PHOTO 2

Photo 2. Devers substation, showing relatively tall porcelain-glass insulators on 10-ft perches with little lateral support. Insulators built in 1982 were destroyed by the July 8, 1986 earthquake.

PHOTO 3

Photo 3. Close-up view of the 1982 transformers (Devers substation).

PHOTO 4

Photo 4. These insulators had very little lateral support---most of the add-on earthquake bracing failed.

Two of the Metropolitan Water District's five pumping stations were shut down due to mechanical damage. This caused the aqueduct to overflow and more than 975 million gallons of water spilled onto the desert floor.

In recent years the Division of Mines and Geology has been increasingly concerned with the effects of earthquakes on lifelines. DMG geologists have prepared planning scenarios for magnitude 8.3 earthquakes on the northern segment of the San Andreas fault (Davis and others, 1982a) and the southern segment of the San Andreas fault (Davis and others, 1982b), and for a magnitude 7.5 earthquake on the Hayward fault (Steinbrugge and others, 1986). In these scenarios an estimate was made of the degree of damage to be expected for various lifeline facilities during large earthquakes based on seismic intensity distribution maps. Smaller earthquakes, such as the one at North Palm Springs, are only modest tests of the prognostications. This article explores some of the details of the disruption of power following the July 8 event.

GEOLOGY AND TECTONICS OF SITE

The Devers substation lies in the Coachella Valley, a broad alluvial plain between the north and south branches of the San Andreas fault (Proctor, 1968) (Figure 1). It was 7.6 km from the epicenter of the earthquake and about 200 m north of the south branch of the San Andreas fault (also known as the Banning fault at this locality) (Figure 2).

During the earthquake, minor surface faulting occurred along a 4-5 km segment of the fault centered near Highway 62 (Figure 2) where approximately 7 cm of right-lateral displacement was observed across a 145 m wide area of cracks. This surface rupture was discontinuous and decreased away from the highway. The most easterly ruptures ended as a wide zone of minor cracks just south of Devers substation. The 180 m wide zone of cracks observed in Diablo Road presumably reflects both tectonic deformation and intense shaking along the fault.

FIGURE 2

Figure 2. Alquist-Priolo Special Studies Zones in the vicinity of the Devers substation. Selected ground rupture data from the North Palm Springs July 8, 1986 earthquake are also shown.

Devers substation is built on Holocene alluvium (sand and gravel) which overlies thick Pleistocene fan deposits (Procter, 1968; Rasmussen, 1981). The depth to ground water reportedly exceeds 18 m (60 feet) at a site slightly lower in elevation and less than 1 km to the southwest of the substation (Rasmussen, 1981). However, the subsurface conditions at the substation are not well known. No evidence of ground failure (cracks or liquefaction) was noted at the substation.

Strong-motion instruments are used to record peak accelerations in both horizontal and vertical directions above a pre-set threshold value. The nearest Division of Mines and Geology strong-motion instrument was at the Desert Hot Springs fire station about 10 km from the epicenter (Figure 1). This instrument measured accelerations of 0.59 g vertical and 0.33 g horizontal (Huang and others, 1986). Despite these high accelerations, there was no structural damage to the firehouse. Pictures, plaques, and helmets hanging on small, 1/2-inch long nails remained in place. A few unriveted ceiling panels fell, books fell off shelves, and dishes fell out of cabinets. These shaking effects may be classified as Modified Mercalli intensity VI.

PHOTO 5

Photo 5. Cracks in adobe walls of recently-remodeled house in Whitewater Canyon. This is indicative of intensity VII on the Modified Mercalli scale.

Along the fault to the east-southeast, a U. S. Geological Survey strong-motion instrument at North Palm Springs, about 10 km from the epicenter, registered a vertical acceleration of 0.78 g and a horizontal acceleration of 0.68 g (Nilson and others, 1986) (Figure 1). Along the fault to the west-northwest, most residences in the Painted Hills subdivision lost all loose and hanging articles to the floor. In addition to the severe cracking of adobe walls in a house in Whitewater Canyon (Photo 5), a few chimneys fell (Photo 6). Thus, while the maximum Modified Mercalli intensity of the event was VII, the peak vertical and horizontal accelerations at the substation must have been greater than 0.78 g and 0.68 g, respectively.

POWER LINES

Electrical power enters the region via a 500,000 volt transmission line from the Palo Verde Nuclear Power Plant in Arizona. After the earthquake, all of the high towers and high voltage lines remained intact. Only local powerlines within a few hundred meters of the surface fault rupture were damaged in any way. These power lines were located on the edges of ridge crests where seismically induced landslides or slumps occurred. The resulting movements of the utility poles were relatively minor, and the lines remained in service.

THE SUBSTATION

Except for power plants, substations remain the most critical components of the electrical delivery system. As power requirements increase, it will be necessary to increase the number of transmission lines or increase the voltage carried on each line, or both. Extremely high voltages, such as those arriving at the Devers substation, put severe restrictions on the design of equipment.

PHOTO 6

Photo 6. A few chimneys fell in Whitewater Canyon. This is indicative of intensity VII on the Modified Mercalli scale.

At Devers, equipment installed in 1967 performed quite well (Photo 1), while that installed in 1982 was largely destroyed (Photo 2). The 1967 transformers had relatively short insulators which were built at an angle, thus giving them lateral as well as vertical support. The 1982 transformers had relatively tall porcelain-glass insulators on 3 m high perches (Photo 3). These insulators had very little lateral support-most of the add-on earthquake bracing failed (Photo 4).

As with other types of heavy equipment, the anchorage of transformers is most important. In one design, a 650,000-pound transformer was sheared from its retaining bolts and moved 10 inches to the south. In another, anchoring clamps for a similar transformer were bent, but the transformer did not move enough to damage other equipment.

Working around the clock, crews took nine days and one hour to energize the 500,000 volt system to the Perris-Hemet area. On the 10th day, repairs to the Devers substation were completed and the system could then transmit power from Arizona. Replacement components had to be shipped from a central store yard as well as from other stations in the Southern California Edison Company system. Some time was saved by dismantling and moving an entire breaker from a substation with less demand.

CONCLUSIONS

The severe damage at the Devers substation was caused entirely by ground shaking from the earthquake on July 8, 1986. No signs of liquefaction, fault rupture, or other permanent ground movement were detected in the substation proper. The nearest ground breakage occurred along the southern branch of the San Andreas fault about 200 m south of the substation. This consisted of hairline fractures with apparent right-lateral displacement totaling not more than 1.6 cm as of July 10th.

Although high voltage transmission lines survived in serviceable condition, the Devers substation---the critical link in the power network---did not. The unexpected heavy damage to recently built transformers and insulators at this substation will stimulate improvements in earthquake-resistant design for handling today's high voltage requirements. It will also be an encouragement to stockpile replacement equipment on-site. The example of Devers substation provides a valuable stimulus for reevaluating the design of other substations located especially close to active faults.

ACKNOWLEDGMENTS

John Kelley, Manager of the Devers substation, interrupted his busy schedule to give us a very informative tour of the substation on the day after the earthquake. We wish to thank John Kelley and Karl V. Steinbrugge, structural engineer, El Cerrito, and Earl Hart, DMG, for reviewing the manuscript.

REFERENCES

California Division of Mines and Geology, 1980, revised, Special Studies Zones map of the Desert Hot Springs quadrangle, scale 1:24,000.

Davis, J. F., Bennett, J. H., Borchardt, Glenn, Kahle, J. E., Rice, S. J., and Silva, M. A., 1982a, Earthquake planning scenario for a magnitude 8.3 earthquake on the San Andreas fault in southern California: Division of Mines and Geology Special publication 60, 128 p.

Davis, J. F., Bennett, J. H., Borchardt, Glenn, Kahle, J. E., Rice, S. J., and Silva, M. A., 1982b, Earthquake planning scenario for a magnitude 8.3 earthquake an the San Andreas fault in the San Francisco Bay Area: Division of Mines and Geology Special Publication 61, 160p.

Huang, N. J., Sherburne, R. W., Parke, D. L., and Shakal, A. F., 1986, California strong-motion records from the Palm Springs, California earthquake of 8 July 1986: Division of Mines and Geology, Office of Strong Motion Studies Report No. 86-5, 74 p.

Jennings, C. W., 1977, Geologic map of California: Division of Mines and Geology, Geologic Data Map No. 2, scale 1:750,000.

Kahle, J. E., Borchardt, G., Manson, M. W., and Hart, E. W., Unpublished field notes of the 1986, North Palm Springs earthquake of 8 July 1986.

Nilson, J. and others, 1986, Strong-motion accelerograms of the southern California earthquake of July 8, 1986: A quick look: U. S. Geological Survey Memo, 6 p.

Proctor, R. J., 1968, Geology of the Desert Hot Springs-Upper Coachella Valley area, California: California Division of Mines and Geology Special Report 94, 50 p.

Rasmussen, G. S., 1981, Engineering geology investigation [of a] 1,171 foot (east-west) by 2,700 (north-south) rectangular parcel, tentative Tract No. 16847, Lots 1-67, North Palm Springs, California: Gary S. Rasmussen and Associates, San Bernardino, California, unpublished consulting report prepared for Robert A. Hammack Development Company, GSR Project No. 1697, 14 p. plus appendices.

Steinbrugge, K. V., Lagorio, H. J., Davis, J. F., Bennett, J. H., Borchardt, Glenn, and Toppozada, T. R., 1986, Earthquake planning scenario for a magnitude 7.5 earthquake on the Hayward fault, San Francisco Bay area: CALIFORNIA GEOLOGY, v. 39, no. 7, p. 153-157.