by S.W. Swan, P.J. Flores and J.D. Hooper
Following the Manzanillo earthquake of October 9, NCEER supported a reconnaissance team led by EQE Inc. to investigate damage. This article is taken from a longer paper of the same title. For more information, contact Paul Flores, EQE Inc. at (714) 833-3303. In related research efforts, EQE has another paper which details the performance of Mexico City's earthquake early warning system, which was triggered on September 14, 1995 following a magnitude 7.2 earthquake in the Mexican state of Guerrero. For more information, contact either Paul Flores or James Goltz at the number given above.
On Monday, October 9, 1995 at 9:37 a.m. local time, a magnitude 7.9 earthquake occurred in the subduction zone off the Pacific Coast of the state of Colima, Mexico. Initial reports from the Institute of Geophysics of the National University of Mexico (UNAM) located the earthquake's epicenter just offshore about 30 kilometers southeast of the port of Manzanillo (18.7oN, 104.3oW with a depth of 33 km). As of October 19, 1995, there had been approximately 50 aftershocks of M>4.0, the largest occurring on October 19, 1995 and registering M=6.1.
The Manzanillo earthquake is the largest and most destructive to occur in the Jalisco region since the magnitude 8.2 event of June 3, 1932. The 1932 event and a magnitude 7.8 aftershock ruptured the Rivera subduction zone, which accommodates downward thrusting of the Rivera plate beneath the Jalisco region as shown in figure 1 (Eisler and McNally, 1984). Since 1932, the Rivera subduction zone had been relatively aseismic (DeMets et al., 1995), indicating that the subduction fault was locked. The October 9, Manzanillo earthquake apparently relieved the strain that had built up during the last 63 years.
This type of subduction movement along the Pacific coastal region is the usual cause of large destructive earthquakes in Mexico. The subduction belt extends along the Pacific coastline from Jalisco south around the southern end of Mexico past the states of Michoacan, Guerrero, and Oaxaca to the Isthmus of Tehuantepec. Seismologists have postulated sections along this 2,000 kilometer belt as "seismic gaps" - sections where large earthquakes have not occurred in decades and are considered overdue (see figure 2). The Guerrero Gap is the most well known and is an anticipated source of large earthquakes that could damage Acapulco and distant Mexico City. There are actually six possible seismic gaps that have been postulated, the Tehuantepech Gap near the isthmus, the Oaxaca Gap, the San Marcos Gap, the Guerrero Gap, the Michoacan Gap, and the northernmost Jalisco Gap. The Michoacan Gap was eliminated by the magnitude 8.1 earthquake in 1985, which caused massive damage in central Mexico City, and according to a preliminary report from the GPS Collaborative Project (UNAM - CalTech, University of Wisconsin, Madison - University of California, Berkeley), the October 9, 1995 Manzanillo earthquake appears to have eliminated the Jalisco Gap.
There appears to have been only one recording of strong motion in the high intensity region of the Manzanillo earthquake. In the late 1980s, the Electric Power Research Institute (EPRI) of Palo Alto, California sponsored the installation of accelerograph systems at four different sites along the southern coast of Mexico. The large oil-fueled power generating plant adjacent to the town of Manzanillo is the northernmost site of the four EPRI accelerograph installations. The Manzanillo Power Plant accelerograph systems were installed and are operated by engineers from the Centro de Instrumentacion y Registro Sismico (CIRES), a non-profit company headquartered Mexico City. The CIRES organization also installed and operates an earthquake early warning system for Mexico City.
On-site analysis of accelerograph records by CIRES engineers characterized the ground motion at the Manzanillo Power Plant site (see figure 3). The CIRES free field accelerograph recorded strong motion (cycles of acceleration on the order of 0.10g or greater) that lasted for a period approaching 30 seconds in the main shock of Monday, October 9. In the most intense period recorded, there were several cycles of motion on the order of 0.40g. The amplitude of ground acceleration appeared to be comparable in the horizontal and vertical directions.
As the intensity of ground shaking can vary considerably even above the fault plane rupture, it is desirable to have multiple recordings of ground motion in earthquakes. Unfortunately, the power plant recording appears to be the only record of motion in the severely shaken region of the earthquake. The observed effects at the Manzanillo Power Plant however indicate that the CIRES record is probably a good representation of the intensity of ground shaking within areas where the earthquake damage was serious. The power plant is located on a low-lying island consisting of a shallow layer of sand, alluvium and artificial fill over bedrock, located between the Pacific shoreline and a lagoon that separates the plant from the town of Manzanillo. Soil conditions at the power plant would be described as moderately compact to soft, probably comparable to soil conditions in the damaged areas further up the coast. The adjacent town of Manzanillo, however, appears to rest on rock or shallow compact alluvium over rock. The earthquake effects and therefore the apparent intensity of ground shaking were much milder within Manzanillo. It is likely that the intensity of ground motion in the town was on the order of half the intensity of shaking measured at the power plant.
The earthquake caused moderate to serious damage over a region extending some 200 kilometers between Manzanillo and Puerto Vallarta (see figure 4). It is estimated that more than 50 people died in the earthquake and over 10,000 were left homeless. The earthquake is notable for directing damage northwestward from the epicenter, up the Pacific coastline, while sparing the epicentral region and the area down the coastline.
Within the Manzanillo city center, damage was minor despite proximity to the epicenter and the concentration of unreinforced masonry (URM) structures. Damage was also relatively light down the coastline toward the epicenter, and inland within the state capital city of Colima. Alternately, serious effects from the earthquake occurred northward in the hotel district of Manzanillo, up the coastline from the city center, and in the smaller communities toward Puerto Vallarta.
The major population centers in the small state of Colima are the capital city of Colima and the primary port of Manzanillo, both with populations of about 130,000. Manzanillo includes the largest industrial port between Lazaro Cardenas on the Michoacan-Guerrero border to the south and Puerto Vallarta to the north. The port of Manzanillo is a modern container facility and a primary shipping and import center for the Pacific coast. Although a smaller tourist center than Puerto Vallarta, the Manzanillo hotel district includes a few mid-rise hotel and condominium complexes, located north of town along the bay shoreline.
In the larger state of Jalisco, the earthquake-affected area extended from the Colima-Jalisco border northward to Puerto Vallarta. All of the coastal towns are small farming and/or fishing communities, with populations of less than 20,000, with the exception of Puerto Vallarta. The bay of Puerto Vallarta includes some two dozen concrete frame high-rise hotels lining the beach front, primarily constructed in the 1970s and 1980s.
The absence of serious effects in the epicentral region may be the most remarkable aspect of the Manzanillo earthquake. The most significant damage was observed from northern Manzanillo to the Colima-Jalisco border.
Nearest the epicenter are the small towns of Ameria and Tecoman, with the state capital of Colima further to the east. Effects from the earthquake were found to be minimal in these communities; corresponding to Modified Mercalli Intensity (MMI) VI to VII. Minor cracking was observed in plaster covering masonry. Abutting buildings sometimes showed minor signs of pounding, such as small cracks and spalling at adjacent walls. The only significant effects in the epicentral area appeared to be a fire triggered by boiler explosion in a small fabrication plant near Tecoman.
Indications of increased shaking intensity begin to appear approaching Manzanillo from the downcoast epicentral area. Small rock slides were observed along the highway, partially blocking the lane adjoining the road cut in a few places. Concrete bridges displayed minor settlement at their abutments. These effects were not sufficient to limit traffic flow.
The central district of Manzanillo is located at the southeast end of the large bay that also includes the industrial port and the hotel district. The foothills of the nearby mountains extend into Manzanillo's center. Most of the town site appears to be essentially founded on rock. The waterfront likely consists of sand fill inside the seawall, as opposed to the firm site conditions underlying the rest of the town. Observed effects in the city center correspond to MMI VI to VII. Masonry infill walls and URM structures suffered minor cracking. One multistory structure displayed diagonal cracks in its infill, characteristic of significant damage to shear walls.
The adjacent industrial port reported minor damage due to settlement and spreading beneath the gantry crane rails. The port includes two large cranes for off-loading container ships. The cranes were reportedly unaffected, but lost about 30% of their range of travel due to settlement beneath the rails.
The Manzanillo Power Plant is sited adjacent to the town on the shoreline at the southern end of the bay (see figure 5). A lagoon about half a kilometer in width separates the plant from the city center. Soil conditions for the shoreline spit, which include the plant, were not available, but the site appears to consist of shallow layers of alluvium and sand over bedrock. The most serious effects at the plant were caused by liquefaction-induced spreading and settlement near the cooling water intake (see figure 6). The siting of the plant between the lagoon and the Pacific was done to provide a large source of cooling water. This requires the cooling water intake to be sited on fill behind a seawall, where it was subject to the same settlement experienced in the industrial port. Settlement near the intake severely damaged the buried, reinforced concrete circulating water lines.
The hotel district of Manzanillo and the surrounding residential areas extend around the side of the bay. This area lies on a narrow coastal plane between the foothills Sierra and the shoreline, which includes marshland. An abrupt difference was observed in earthquake effects between the city center located on firm soil to the south and the hotel and residential district on the coastal plain to the north. Effects around the hotel district generally correspond to MMI VIII.
The hotel district included the earthquake's single disastrous collapse, the six-story, infilled concrete-frame Costa Real Hotel, located on the beach at about the bay's midpoint (see figure 7). According to articles published in local newspapers, the hotel was seriously damaged by long distance motion in the magnitude 8.1 earthquake of 1985. The hotel was closed for about two years following the 1985 earthquake and then reopened under new ownership. Allegedly, repairs were never made to fractures in the concrete framing and infill, other than simple cosmetic plaster. The hotel suffered a complete "pancake" collapse of all floors, and accounted for 38 of the estimated 50 fatalities in the earthquake.
In the same vicinity, structures similar to the Costa Real Hotel suffered varying degrees of damage, including separations at the interface of infill walls and the enclosing framing, loss of masonry at corners and from parapets, and limited damage to concrete frames. Several one- and two-story structures experienced partial or total collapse in the hotel district. These included two small buildings on the same grounds as the Costa Real Hotel, a possible indication that the site might have been a focus of particularly strong motion.
Damage to the main hospital in Manzanillo was another major source of disruption (see figure 8). The hospital, not far from the Costa Real, is a four-story concrete shear wall structure constructed in the early 1980s. The building experienced moderate cracking in shear walls and shear cracking in two interior columns at the main level and in 10 exterior columns at the mechanical penthouse. Damage appeared to result from "short column" effect, where adjoining partial-height walls force columns to deform in shear, rather than bending - a more brittle and damaging mode. Initally, damage to the hospital was considered sufficiently serious to prompt evacuation of all patients. A more detailed examination of the building indicated that it was not in danger of collapse.
Another disastrous structural failure occurred in a small shopping center in the village of Santiago (see figure 9), located just north of Manzanillo. The exterior masonry cladding toppled from the second floor of a two-story, infilled concrete-frame market building. A number of people were shopping in the mall area below; a total of eight were killed by falling masonry.
In general, the performance of lifelines in the affected area was good. Long-term interruptions of lifeline services did not occur nor did the damage experienced result in loss of life, impairment to public health and safety, or apparent serious economic loss to the region.
The societal impacts from the earthquake were substantial. Approximately 10,000 persons were displaced from their homes in the cities of Manzanilllo and Cihuatlan and the small fishing port Barra de Navidad. Some people continued to live in their homes even though they were visibly unsafe to occupy. Except for some large facilities like the Social Security hospital in Manzanillo and the municipal administration hall in Cihuatlan, very few damaged structures had been identified as having been inspected for safety purposes.
Many of the displaced people were not in official shelters set up by either the government or the Mexican Red Cross. There appeared to be several spontaneously established shelters in sport fields or other open space areas where people had set up makeshift shelters using tents or other materials. It was difficult to assess how many people were actually displaced by the earthquake damage since according to newspaper accounts, many refused to reoccupy their homes as long as the aftershock activity continued.
Another concern expressed by many is the impacts that the earthquake might have on the tourist industry in the area. While the direct effects on the tourist industry and its infrastructure (hotels, roads and highways, airports, etc.) were minimal, there was some concern expressed that tourists from the United States or elsewhere might cancel their trips because of the perception of continuing seismic activity along Mexico's west coast. The tourist season in Manzanillo and the surrounding areas begins in December. Without some formal study that can measure this type of economic impact, it is difficult to reach any conclusion about the impacts on the tourist industry at this time.
In the City of Manzanillo, the most pressing emergency response issue was the urban search and rescue efforts at the site of the collapsed Costa Real Hotel. A Japanese search and rescue team was assisting municipal civil protection personnel with the effort. According to a rescuer on site, only three victims were pulled out from the debris still alive on the same day of the earthquake. The rescue operation was stopped on Friday, October 13, at 2:00 p.m. and debris removal began soon after.
Emergency medical assistance did not seem to present a problem even though the Social Security Hospital was evacuated due to structural damage. Forty patients were receiving inpatient care at the time of the earthquake and had to be transported to other medical facilities as far away as the City of Colima. According to press accounts, the director of health services for the state of Colima indicated that the closure of the hospital and other important medical clinics required that a medical emergency be declared and plans be put into action to ensure that ongoing medical care was available to the population of the area.
DeMets, C., Carmichael, I., Melbourne, T., Sanchez, O., Stock, J., Suarez, G. and Hudnut, K., (1995), "Anticipating the Successor to Mexico's Largest Historical Earthquake," EOS, Vol. 76, No. 42.
Eisler, H.K. and McNally, K.C., (1984), "Seismicity and Tectonics of the Rivera Plate and Implications of the 1932 Jalisco, Mexico, Earthquake," Journal of Geophysical Research, Vol. 89, 4520.
Some of the material reported herein is based upon work supported in
whole or in part by the National Science Foundation, the State of New York, the U.S.
Department of Transportation, the Federal Emergency Management Agency and other sponsors.
Any opinions, findings, and conclusions or recommendations expressed in this publication
are those of the author(s) and do not necessarily reflect the views of NCEER or its
Click here to return to the Table of Contents
| Home | NCEER | Information Assistance | Publications | Databases, Software |
| Events | New! | Links | Data Resources |