A Comparison of
ARTIFICIAL AND NATURAL SLOPE FAILURES
THE SANTA BARBARA EARTHQUAKE OF AUGUST 13, 1978
EDWIN L. HARP, DAVID K. KEEFER, AND RAYMOND C. WILSON
Geologists. U.S. Geological Survey, Menlo Park, CA
The August 13,1978, earthquake (M, = 5.1, Caltech Seismological Laboratory; M = 5.5, National Earthquake Information Service, U.S. Geological Survey, Golden, Colorado; ML = 5.7, Berkeley Seismographic Station) occurred 4 km offshore south of Santa Barbara at 3:54 p.m. local tiine (figure l). Significant local damage occurred in Santa Barbara and on the campus of the University of California at Goleta 15 km to the west (Miller and Felsseghy, 1978). Scores of people were treated for minor injuries. Seismological investigations of the earthquake and its aftershocks (Lee and others, 1978) indicate that the earthquake was caused by reverse faulting on a north- dipping fault at approximately 12 km depth, and that rUpture propagated northwest toward Goleta.
The earthquake triggered rockfalls and rockslides from steep road cuts and coastal cliffs. One of the largest rockfalls (about 100 cubic meters) closed California State Highway 154, 5 km southeast of San Marcos Pass for approximately 30 hours. Extensive ground cracking occurred in a roadside turnout about 4.5 km southeast of San Marcos Pass and several hundred cubic meters of rockfall debris was shaken from the steep slopes beneath the turnout.
Other small rockfalls (generally less than 10 m3 in volume) occurred in roadcuts along Highway 154 in the San Marcos Pass area and along U.S. Highway 101 south. Slumping and settlement of a railroad embankment near Ellwood, about 10 km west of Santa Barbara, caused a train derailment that occurred approximately seven minutes after the earthquake. Aside from rockfalls from steep slopes below the roadside turnout 4.5 km southeast of San Marcos Pass, the only natural slope failures from this earthquake were observed in the coastal cliffs near Santa Barbara and Goleta Point where small rockfalls occurred.
The Santa Barbara region is an active seismic area and has been subjected to numerous moderate and strong historic earthquakes (Hamilton and others, 1969; Lee and Ellsworth, 1975; Lee and others, 1978). The December 21, 1812, earthquake was estimated to be magnitude 7 1/4. Much of Santa Barbara was damaged by the magnitude 6.3 earthquake of June 29, 1925. Another strong earthquake of magnitude 6.0 occured June 30, 1941, which also caused damage in Santa Babara (Lee and otheres, 1978).
The August 13, 1978 Santa Barbara earthquake affordeand opportunity to gather data on landslides resulting froma moderate earthquake. A reconnaissance of landslides from this earthquake was undertaken to determine (1) the types of landslides produced, (2) the landslide distribution, and (3) to assess the hazard posed by the landslides.
RECONNAISSANCE SURVEY OF LANDSLIDES
The landslide survey was accomplished by traverses across the Santa Ynez range along State Highway 154, U.S. 101, and several local side roads; by traverses along the north and south flanks of the Santa Ynez range; and by examination of the coastal bluffs near Goleta Point (figure 1).
San Marcos Pass Area
Site 1.A rockfall of about 100 m3 volume occurred along a steep roadcut, 50 m in height, in conglomerates and weakly cemented sandstones of the Tertiary Sespe Formation (figure 1). The debris from this rockfall closed California State Highway 154 for about 30 hours. Although the volume of debris was not large, the rock forming the steep slopes above the road was severely shaken and loosened (figure 2), causing a hazardous condition. California State Highway Department (Cal Trans) blasted the dangerously loose rocks and most of the closure period was due to debris removal from the slope above the road.
Site 2. Approximately 0.5 km north of site 1, rockfalls occurred on 60° slopes below a roadside turnout. The turnout is constructed from fill derived from the adjacent roadcut in Coldwater Sandstone (Dibblee, 1966). The fill is 1 m thick or less and rests upon the slope composed of heavily jointed buff-colored sandstone and siltstone. Below the road at this turnout, the slopes shed several hundred cubic meters of rockfall debris during the earthquake. The thickness of slopes removed by rockfalls was generally less than 1 m and the resulting scarps were spaced over the slope in rather discontinuous areas for a horizontal distance of about 100 m (figure 3).
The fill forming the surface of the turnout was extensively cracked, with cracks near the edge of the turnout showing separations of up to 3.8 cm (figure 4). The cracks extended back from the edge of the turnout for about 6 m. A little over 1 cm of subsidence was observed across several of the cracks closest to the edge. The extent of cracking in this turnout suggests that much of it would have failed if the ground motion had been stronger or of longer duration. Because of these cracks, the site is now less stable.
Site 3. A small rockfall occurred in fault gouge in a roadcut on the Painted Cave Road about 2 km north of Highway 154 (figure 5). The steeply dipping fault surface with well developed horizontal slickensides is exposed in the roadcut. The gouge material had apparently undergone slumping prior to the earthquake as evidenced by the road bed having been entirely resurfaced directly below the exposed fault gouge.
Aside from the above specific sites, other small rockfalls, rockslides, and cracks were observed in roadcuts within the area along Highway 154 from 2.5 km north of Santa Barbara to about 3 km north of San Marcos Pass (figure 1). These failures consisted of one or several rocks or small piles (generally less than 1 m3 volume) of gravel-sized debris which had fallen from roadcuts. Almost all of these failures were from weakly cemented conglomerates of the Sespe Formation or fractured, silty sandstones of the Coldwater Sandstone.
Small rockfalls and rockslides occurred in roadcuts along the Painted Cave and East Camino Cielo Roads north and east of Highway 154 (figure 6) as a result of boulders being dislodged from roadcuts by the shaking. All of the failures on the Painted Cave and East Camino Cielo Roads occurred in Coldwater Sandstone.
Gaviota Pass Area
Several small rockfalls and soilfalls less than 1 cubic meter in volume were noted in a limited area along U.S. 101 near Gaviota Pass (figure l). These failures were in sandstone and siltstone of Tertiary Gaviota and Sacate Formations (Dibblee, 1950). All failures observed in this area were in roadcuts; no failures were seen in natural slopes.
Site 4. Steep ocean cliffs comprised of diatomaceous shale of the Sisquoc Formation (Dibblee, 1966) at and near Goleta Point on the campus of the University of California at Santa Barbara produced numerous small (1 m3 or less) rockfalls that were deposited in scattered piles of gravel-sized particles and occasional fragments up to boulder size along the base of the cliffs (figure 7). Numerous small rockfalls were reported to the east along the ocean cliffs south of Santa Barbara and west to Goleta in slopes of Monterey shale (R. Yerkes, U.S. Geological Survey, 1978, oral communication). The small particle size of the rockfall debris is due to the close spacing of fractures and the thin-layered bedding in the shale of the Sisquoc and Monterey Formations.
Site 5. Slumps and cracks occurred in a railroad embankment near Ellwood (figure 1). The embankment material is an uncompacted mixture of excavated rock, soil, and waste lumber. Vibrational consolidation of the fill during the earthquake was reportedly responsible for the derailment of a Southern Pacific freight train (figure 8) passing this area approximately seven minutes after the earthquake (J. Newby, soils engineer, Southern Pacific Railroad, 1978, oral communication). The seismic-induced settlement of the embankment reportedly affected the alignment of the railroad tracks sufficiently to derail one diesel unit and 20 cars of the passing freight train (Los Angeles Times, August 14, 1978). At the time of this reconnaissance, the second day after the earthquake, all traces of cracks, slumps, and settlement had been eradicated and the track relaid through the section torn during the derailment. However, several other cracks and small slumps in the embankment were observed just east of the derailment area (figure 9). Similar failures of railroad embankments were numerous during the 1925 earthquake along the Southern Pacific Railroad west of Santa Barbara, where settlement of 8 inches to 3 feet in railroad embankments were reported (Kirkbride, 1927).
GENERAL SLOPE RESPONSE
The earthquake of August 13 was apparently just above the threshold of ground motion necessary to generate landslides in this region. The predominant types of landslides generated from this earthquake are similar to those of landslides generated by other moderate earthquakes in California (Fortuna-Rio Dell, 1975, M = 5.3, (Harp, 1975); Oroville, 1975, M = 5.7 (Harp and Wilson, 1975). Preliminary intensity surveys (C. Stover, U.S. Geological Survey, 1978, oral communication) indicate that the threshold of rockfall generation in the Santa Barbara earthquake corresponded to a shaking intensity of Modified Mercalli intensity V to VI. Because shaking levels necessary to cause rockfalls were apparently barely exceeded in this earthquake the slope failures which occurred provide a comparison of the seismic behavior of artificial and natural slopes. The pattern of landslide occurrence indicates that roadcuts are more susceptible to seismic-induced failure than are natural slopes; no failures were observed from natural slopes except along the steep coastal bluffs and at site 2 in the San Marcos pass area.
SUMMARY AND CONCLUSIONS
Although the landslides from the August 13, 1978, earthquake were not widespread nor spectacular in size, their occurrence was instructive in pointing out some of the existing and possible future hazards from earthquake-induced land-slides in the Santa Barbara area:
(1) Most rockfalls and rockslides occurred in steep roadcuts.
(2) The steepness of the roadcuts, close fracture spacing of the rocks, and weak cementation were the most important factors in the formation of rockfalls and rockslides.
(3) Natural slopes generally were not susceptible to rock falls except along the ocean cliffs.
(4) The presence of closely fractured shales and near-vertical slopes were two important factors influencing rockfall occurrence along ocean cliffs.
(5) The sites of rockfalls and rockslides in this earthquake will be sites susceptible to seismic-induced landslides from future earthquakes; these sites will also show increased susceptibility to non-seismic slope failures because of the shaking-induced loosening of rocks in fractured roadcut slopes, especially at sites 1 and 2.
(6) The failure of the railroad embankment near Ellwood probably was due in large part to the uncompacted nature of the fill. The susceptibility of uncompacted fills to seismic-induced failures and the record of similar failures in the 1925 earthquake suggest that such failures will pose hazards to the railroad in future earthquakes.
Despite the moderate size of this earthquake and the small size of most of the landslides, lifelines were closed and significant expense was incurred as a result of seismic-induced ground failure. In future earthquakes, problems of lifeline access are again likely to occur from landslide damage. These problems are likely to present the greatest hazard to the railroad and to people who live in the Santa Ynez Mountains and use Highway 154 or other mountain roads as access to supplies and medical facilities.
Dibblee. T.W. Jr.. 1950. Geology of southwestern Santa Barbara County. California: California Division of Mines. Bulletin 150.84 p.
Dibblee. T.W. Jr.. 1966. Geology of the central Santa Ynez Mountains. Santa Barbara County. California: California Division of Mines and Geology. Bulletin 186.99 p.
Hamilton. R.M.. Yerkes. R.F.. Brown. R.D.. Jr.. Burford. R.D. and De Noyer. J.M.. 1969. Seismicity and associated effects. Santa Barbara region. pt. D in Geology. petroleum development. and seismicity of the Santa Barbara Channel region. California: U.S. Geological Survey Professional Paper 679. p. 47-68.
Harp. E.L.. and Wilson. R. C. 1975. Reconnaissance report on slope failures from the 1 August 1975 earthquake at Oroville California: Report on file in Menlo Park. U.S. Geological Survey.
Harp. E.L.. 1975. Reconnaissance of ground failure and landslide effects of the June 7. 1975. earthquake near Fortuna, California. in Investigations of the June 7. 1975. earthquake in Humboldt County. California: U.S. Geological Survey Open File Report 75-404. p.5 -10.
Kirkbride. W.H.. 1927. The earthquake at Santa Barbara. California. June 29. 1925. as it affected the railroad of the Southern Pacific Company: Seismological Society of America Bulletin. v. 17. no. 1. p. 1-7.
Lee. W.H.K.. and Ellworth. W.L.. 1975. Earthquake activity in the Santa Barbara Channel region in Draft Environmental Statement. Oil and gas development in the Santa Barbara Channel outer continental shelf off California: U.S. Geological Survey. P. IIl80-II-137.
Lee. W.H.K.. Johnson. C.E.. Henyey. T.L.. and Yerkes. R.F. 1978. A preliminary study of the Santa Barbara earthquake of August 13.1978. and its major aftershocks: U.S. Geological Survey Circular No. 797. 11 p.
Los Angeles Times. August 14. 1978.
Miller. R.K.. and Felszeghy. S.F.. 1978. Engineering features of the Santa Barbara earthquake of August 13. 1978: Earthquake Engineering Research Institute Special Report UCSB-ME-78-2.