5
Lifeline Perspective

Ronald T. Eguchi and Hope A. Seligson

ABSTRACT

This paper provides a summary of important lessons learned after the Loma Prieta earthquake. As a result of this event, a comprehensive research agenda was sponsored by the National Science Foundation (NSF) in collaboration with the other National Earthquake Hazard Reduction Program (NEHRP) agencies. In total, over 80 research projects were funded addressing seismological, geo-technical, structural, lifeline, and socioeconomic topics. This particular paper summarizes research findings resulting from studies on lifeline performance. Lifelines considered in this paper include water supply, wastewater, natural gas, oil, electric power, and communication systems. Transportation systems and port and harbor facilities are addressed in companion papers.

INTRODUCTION

After every major earthquake, there is a ''window of opportunity'' for researchers to advance the science of earthquake engineering. In cases where practical lessons are learned, these opportunities can result in significant changes in seismic analysis, design, and construction procedures. In order to substantiate these lessons, however, comprehensive, well-focused research is needed.

The NSF, along with the other NEHRP agencies, has been sponsoring earthquake-specific research since the 1971 San Fernando earthquake. In addition to the San Fernando earthquake, research initiatives were established after the 1985 Mexico City earthquake, the 1987 Whittier Narrows earthquake, and the 1989 Loma Prieta earthquake. Research agendas for these initiatives usually focused



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 135
Practical Lessons from the Loma Prieta Earthquake: Report from a Symposium Sponsored by the Geotechnical Board and the Board on Natural Disasters of the National Research Council 5 Lifeline Perspective Ronald T. Eguchi and Hope A. Seligson ABSTRACT This paper provides a summary of important lessons learned after the Loma Prieta earthquake. As a result of this event, a comprehensive research agenda was sponsored by the National Science Foundation (NSF) in collaboration with the other National Earthquake Hazard Reduction Program (NEHRP) agencies. In total, over 80 research projects were funded addressing seismological, geo-technical, structural, lifeline, and socioeconomic topics. This particular paper summarizes research findings resulting from studies on lifeline performance. Lifelines considered in this paper include water supply, wastewater, natural gas, oil, electric power, and communication systems. Transportation systems and port and harbor facilities are addressed in companion papers. INTRODUCTION After every major earthquake, there is a ''window of opportunity'' for researchers to advance the science of earthquake engineering. In cases where practical lessons are learned, these opportunities can result in significant changes in seismic analysis, design, and construction procedures. In order to substantiate these lessons, however, comprehensive, well-focused research is needed. The NSF, along with the other NEHRP agencies, has been sponsoring earthquake-specific research since the 1971 San Fernando earthquake. In addition to the San Fernando earthquake, research initiatives were established after the 1985 Mexico City earthquake, the 1987 Whittier Narrows earthquake, and the 1989 Loma Prieta earthquake. Research agendas for these initiatives usually focused

OCR for page 135
Practical Lessons from the Loma Prieta Earthquake: Report from a Symposium Sponsored by the Geotechnical Board and the Board on Natural Disasters of the National Research Council on problems or issues uncovered as a result of these events. From this perspective, post-earthquake research has been problem focused. The Loma Prieta earthquake offered a number of unique research opportunities. In the lifeline area, this earthquake allowed a detailed examination of seismic design procedures originally introduced as a result of the San Fernando event. In many cases, failure of lifeline systems was prevented because of these measures; in some cases, new vulnerabilities were uncovered. In general, research has been directed at explaining why certain design or construction measures work and why others do not. Analyzing the earthquake vulnerability of our nation's lifeline systems is critical for several reasons. First, from the standpoint of replacement cost, life-lines account for approximately $4.5 trillion, or roughly 22 percent of the total built environment (Jones, 1993). Protecting these assets during natural disasters deserves special attention. Second, the recovery of cities after major natural disasters will depend in large part on the survivability of lifeline systems. As can be seen today in Florida, full recovery after Hurricane Andrew is slow, due in part to a lack of utility service. Many areas are still without electric power service. Finally, many systems are aged and ready for reconstruction or replacement. Taking advantage of the lessons learned from previous earthquakes offers an opportunity to enhance the seismic resistance of these systems. Identifying practical measures that can be applied to the seismic design, retrofit, and construction of lifeline systems is an essential first step in this overall process. The purpose of this paper is to summarize practical lessons learned from research conducted as a result of the Loma Prieta earthquake. In particular, the emphasis is on research conducted to better understand the behavior of lifelines during earthquakes. While the lifeline area covers many different systems, several of the lessons learned apply to more than one lifeline. The applicability of these lessons to more than one lifeline will depend on whether they are connected or discrete systems. Connected systems generally include those that rely on transmission lines to convey service. Discrete lifelines may be classified as terminal or source facilities, for example, ports, harbors, and airports. In this paper, the emphasis is on connected systems, that is: the water supply; wastewater; natural gas; oil; electric power; and communication. Even in the specialized area of lifeline earthquake engineering, it is difficult to identify all research efforts conducted as a result of an earthquake. For most government-sponsored research, the identification of ongoing efforts can usually

OCR for page 135
Practical Lessons from the Loma Prieta Earthquake: Report from a Symposium Sponsored by the Geotechnical Board and the Board on Natural Disasters of the National Research Council be made by contacting the funding organizations and requesting a list of awards. This procedure was used in the preparation of this paper. The primary organization providing research money to study this earthquake was the NSF. Information on organizations and individuals who were awarded research grants by the NSF was provided by Drs. S.C. Liu and C.J. Astill. This help was greatly appreciated. In addition to the NSF, other organizations have provided some money to study this earthquake. The third section of this paper attempts to highlight these efforts, where known. Other efforts that may have been conducted with the support of private funds are not documented here, unless otherwise noted. Because of the wide variety of lifeline systems, it is impossible for any one individual to list and summarize the research results. For this reason, the authors contacted many individuals to solicit their input in the preparation of this paper. The authors would like to first acknowledge the support of the discussant panel: Donald B. Ballantyne, Kennedy/Jenks Consultants; Professor Thomas D. O'Rourke, Cornell University; Charles Roberts, Port of Oakland; and Steven H. Phillips, Pacific Gas and Electric. The authors would also like to acknowledge several individuals who contributed to parts of this paper or who offered technical advice in writing certain sections. These individuals are Douglas Honegger, EQE International; Sam Swan, EQE International; Professor Anshel Schiff, Stanford University; and Alex Tang, Northern Telecom. In addition, there were many individuals who kindly furnished reports or papers, in a very timely manner. These individuals are: Professor A. H-S Ang, University of California, Irvine; Dr. Jeremy Isenberg, Weidlinger Associates; Professor James O. Jirsa, University of Texas; Professor Barclay Jones, Cornell University; Professor Jamshid Mohammadi, Illinois Institute of Technology; and Stuart Werner, Dames & Moore. To all of these individuals, the authors express a sincere thanks. LIFELINE SYSTEM PERFORMANCE DURING THE 1989 LOMA PRIETA EARTHQUAKE The following section presents brief summaries of lifeline system performance in the Loma Prieta earthquake. These summaries are intended to give an overview of system response, rather than list specific component damages. Table 5-1 lists the utility agencies as surveyed by the Earthquake Engineering Research Institute's reconnaissance report. For each utility, an assessment of the earthquake's impact has been made, based on reported damage, response, and recovery. In addition, available data on water and sewage pipeline failures have been included. These data were taken from a compilation by the Technical Council on Lifeline Earthquake Engineering of the American Society of Civil Engineers (ASCE/TCLEE, 1992).

OCR for page 135
Practical Lessons from the Loma Prieta Earthquake: Report from a Symposium Sponsored by the Geotechnical Board and the Board on Natural Disasters of the National Research Council TABLE 5-1 Impact of Loma Prieta Earthquake on Bay Area Utilities   Utility Agencies Located Within the Affected Areas Impact of Earthquake     Number of Pipeline Repairs as Reported in ASCE/TCLEE Pipeline Database Lifeline   None Minor Major   WATER             Alameda County Water District •     0   Aldercroft Valley Water Company •     0   California Water Service Company •     0   Chemekta Park Water Company •     0   City of Cupertino — Water   •   4   City of Hollister—Water     • 7   City of San Francisco — City Water     • 70   City of San Francisco — AWSS     • 5   City of Santa Cruz—Water     • 78   City of Tracy — Water •     0   City of Watsonville — Water     • 52   East Bay Municipal Utility District —Water     • 133   Idylwild Water Company •     0   Mountain Charlie Waterworks Inc.   •   16   Pajaro Community Services District   •   7   Purissima Hills Water District   •   5   Redwood Estates Mutual Water Company     • 70   Riva Ridge Mutual Water Company   •   1   San Jose Water Company     • 202   San Lorenzo Valley Water District     • 54   Santa Clara Valley Water District   •   5   Scotts Valley Water District   •   2   Soquel Creek Water District     • 31   Sunny Mesa Water District   •   1   Sunny Slope County Water District •     0   Villa Del Monte Mutual Water Company   •   5           748 SEWER/SANITATION             City of Hollister — Sewer   •   0   City of Palo Alto — Regional Wastewater   •   1   City of San Francisco — Clean Water Program   •   5   City of San Jose — Sewer   •   2   City of Santa Cruz — Sewer     • 25   City of Scotts Valley — Sewer     • 0   City of Watsonville — Sewer     • 0   East Bay Municipal Utility District—Wastewater     • 38   Pajaro Community Services District •     5   San Mateo — Sewer (indirect)   •   0   Santa Cruz County Sanitation District     • 11   South Bayside System Authority —Wastewater   •   0   Sunny Slope County Water District —Sewer •     0   Union Sanitary District   •   0           87

OCR for page 135
Practical Lessons from the Loma Prieta Earthquake: Report from a Symposium Sponsored by the Geotechnical Board and the Board on Natural Disasters of the National Research Council   Utility Agencies Located Within the Affected Areas Impact of Earthquake   Number of Pipeline Repairs as Reported in ASCE/TCLEE Pipeline Database Lifeline   None Minor Major   COMMUNICATION           Long Distance:   AT&T   •   —   MCI •     —   US Sprint •     — Local Networks:   Continental Telephone (Gilroy)   •   —   General Telephone (Los Gatos)     • —   Pacific Bell   •   — POWER           PG&E       • — GAS           PG&E       • — Water Supply In general, aqueduct and reservoir facilities were undamaged. No major damage to dam facilities was reported, although several minor cracks on embankments and spillways were noted. Storage tanks were damaged in Los Gatos, San Jose, Los Altos Hills, Watsonville, Sunny Mesa, San Lorenzo Valley, Scotts Valley, and the Santa Cruz Mountains. Pipeline damage was extensive in areas of ground failure, such as San Francisco's Marina District, Santa Cruz, and Watsonville. Disruption lasted as long as two weeks in the harder-hit areas. In San Francisco, isolated damages causing the loss of contents of a 750,000 gallon tank severely impacted the city's fire-fighting capability (Figure 5-1). The flexibility of fire-suppression methods and the availability of the city's fire boat, however, minimized the impact of the tank failure. Concern over possible contamination resulted in four cities in the epicentral area issuing "boil water" notices. The notices were in effect for one day in Los Gatos, three days in Watsonville, and seven days in Santa Cruz and San Lorenzo Valley (EERI, 1990). Wastewater Due to power outages and the lack of backup power at pumping plants, sewage was released into San Francisco and Monterey bays, as well as into the Pacific Ocean. These releases could have been avoided had adequate emergency power facilities been available (EERI, 1990).

OCR for page 135
Practical Lessons from the Loma Prieta Earthquake: Report from a Symposium Sponsored by the Geotechnical Board and the Board on Natural Disasters of the National Research Council FIGURE 5-1 Twenty-two structural fires were reported immediately after the Loma Prieta  earthquake. The worst fire began inside a four-story apartment building in the  Marina District, probably as a result of a leaking gas main. Sewerage facilities were damaged in areas that suffered damage to water systems. This damage is less evident, however, and the lack of water service and subsequent disuse of sewer facilities delayed documentation of damage. Damage had been reported in the city of Watsonville, Scotts Valley, and Santa Cruz. Minor damage was also reported at various regional wastewater treatment facilities (Kennedy/Jenks/Chilton, 1990). Natural Gas Pacific Gas and Electric's (PG&E) natural gas transmission system was virtually undamaged—only two leaks were reported. Both were repaired without customer interruption. However, distribution systems in several areas were severely impacted. Over 1,000 pipeline leaks were reported system-wide, and three low-pressure systems were so heavily damaged that replacement was required. Replacement consisted primarily of insertion of plastic pipe into existing mains and services (Figure 5-2). The distribution system in the Marina District of San Francisco was replaced within one month, at a cost of $17 million. Fifty-one hundred customers were affected. Reconstruction of the Watsonville low-pressure system was complete within three weeks, affecting 166 customers. One hundred and forty customers were impacted in Los Gatos, where main restora-

OCR for page 135
Practical Lessons from the Loma Prieta Earthquake: Report from a Symposium Sponsored by the Geotechnical Board and the Board on Natural Disasters of the National Research Council FIGURE 5-2 Within the Marina District, damage to cast-iron natural gas pipe was so extensive that rather than repairing damaged pipe, new polyethylene pipe was inserted in existing mains and services. tion was accomplished within ten days, and service restoration was complete within a month (Phillips and Virostek, 1990). Total gas system damages have been estimated at $19 million (Matsuda, 1993). One of the most labor-intensive parts of the restoration process was the relighting of services. $7 million was incurred after the earthquake to relight pilots that had been turned off as a result of the earthquake (Matsuda, 1993). Service relights, which were accomplished within ten days, were required by 156,355 customers. The majority of relights resulted from customers unnecessarily turning off their own gas in response to hastily worded media messages, which recommended shutoff without specifying "if you smell or hear gas." At the peak of the relight effort, 1,183 servicemen were utilized. Outside utility companies contributing manpower to the relight effort included Southern California Gas, San Diego Gas and Electric, Mountain Fuel, Sierra Pacific, Northwest Natural Gas, and Washington Natural Gas (Phillips and Virostek, 1990). Oil Refineries And Associated Facilities Most of the region's refineries and tank farms are located along the San Francisco Bay in Alameda and Contra Costa counties. Numerous tanks at soft-

OCR for page 135
Practical Lessons from the Loma Prieta Earthquake: Report from a Symposium Sponsored by the Geotechnical Board and the Board on Natural Disasters of the National Research Council soil sites were damaged, predominantly tanks that were full or nearly full. Typical damage modes included elephant's foot buckling, sometimes associated with loss of contents; damage to associated piping; and uplift of unanchored tank walls. It was reported that all leaks were contained within containment dikes and that no fires resulted (EERI, 1990). Electric Power Primarily as a result of direct damage to transmission substations, 1.4 million PG&E customers lost power following the Loma Prieta earthquake. Power was restored to most of San Francisco within seven hours, and all but 12,000 customers had power within two days (PG&E, 1990). In many cases, power restoration was accomplished by bypassing damaged equipment and operating with reduced levels of circuit protection. Damage to power generation and bulk transmission facilities was estimated to be $19 million, while distribution added an additional $4 million in damages (Matsuda, 1993). Damage was severe in several 500-kV switchyards, including Moss Landing in Monterey Bay and Metcalf in the San Jose area. Seven 500-kV circuit breakers required replacement, at a cost of $700,000 each. These items are not stockpiled by PG&E, and only two came from within the PG&E system. The rest had to be obtained from various sources, including the Tennessee Valley Authority, the Los Angeles Department of Water and Power, and Southern California Edison (PG&E, 1990). Power plant damage was minor, but several plants were forced off-line by substation damage. The rapid loss of power throughout parts of the system was fortuitous in that many distribution systems were no longer energized when damaging situations, such as wrapping of lines, occurred. Communications The most notable impact of the earthquake on telecommunications was the monumental increase in call volume, both locally and worldwide. AT&T reported 27.8 million calls attempted to the 415 and 408 area codes the day after the earthquake. Nine and a half million of these calls were completed, more than double the normal daily volume of 3.5 million. Pacific Bell also reported heavy volume within the Bay Area—80 million calls versus the norm of 55 million. Heavy volumes led to dial tone delays, which in some cases impacted emergency communications (911) activities, and there were several days of service degradation during peak load times (EERI, 1990). Direct damage to telecommunications facilities and equipment was limited. Most difficulties resulted from failure of backup power systems or insufficient backup power capacity.

OCR for page 135
Practical Lessons from the Loma Prieta Earthquake: Report from a Symposium Sponsored by the Geotechnical Board and the Board on Natural Disasters of the National Research Council RESEARCH ACTIVITIES INITIATED AFTER THE EARTHQUAKE Post-earthquake research activities usually fall into one of two categories: federally sponsored research or research funded by private or semi-public organizations. In this paper, the emphasis is on research sponsored by the federal government, that is, the NSF. Other efforts, where publicly acknowledged, are also identified later in the paper. Nsf-Sponsored Research While the risk to lifeline systems in earthquakes is generally acknowledged, it is not necessarily understood. Some research into the causes of damage and disruption of lifeline systems in earthquakes has been completed, but the amount is small relative to many other areas of earthquake hazard mitigation study. Data from the NSF Research Awards Database for earthquake hazard mitigation studies were reviewed for the years 1980-1990 (Seligson et al., 1991). Of the approximately $160 million spent, only about $18 million (11 percent) was spent on lifeline research. As noted in the original review, this figure does not incorporate monies routed by funded organizations (e.g., the National Center for Earthquake Engineering Research) into lifeline research and will, therefore, underestimate actual dollars spent on lifeline studies. However, a consistent comparison between general funding patterns and funding in the post-earthquake environment may be made by looking exclusively at NSF funding. While recent funding levels for lifeline earthquake studies are an average of 11 percent of the total amount spent, this number may increase in the post-earthquake environment. Of the approximately $1.4 million in NSF funds awarded to study the 1987 Whittier Narrows earthquake, 24.5 percent were related to the study of lifelines (based on a title search of the NSF awards database, 1980-1990). Studies of structural performance received the greatest percentage of funds, approximately 37.7 percent of the total. Following the Loma Prieta earthquake, the NSF funded $4.2 million in research through the Loma Prieta Initiative. These funds sponsored various engineering investigations, including studies addressing geotechnical, structural, seismological, and socioeconomic topics. The percentage breakdown by discipline is shown in Figure 5-3. Of this $4.2 million, about 15 percent was spent investigating lifeline issues, while 30.3 percent was spent on structural research. Of the monies spent on lifeline research between 1980 and 1990, the majority has gone to investigating water and transportation system facilities. In general, water lifelines have received about 30 percent of the funding dedicated to lifeline topics. This number may be even larger, as studies of various multipurpose, multi-lifeline components are not included in this estimate. If the monies spent on pipeline and tank research were included, this number might be as high as 60 percent. In addition, 24 percent has gone to transportation studies. The

OCR for page 135
Practical Lessons from the Loma Prieta Earthquake: Report from a Symposium Sponsored by the Geotechnical Board and the Board on Natural Disasters of the National Research Council Figure 5-3 Breakdown of NSF Research in the Loma Prieta Initiative ($4.2 Million Total) remainder is distributed among communications (3.3 percent), electric power (2.4 percent), natural gas (1.6 percent), wastewater (0.6 percent), disaster preparedness and emergency response (3.4 percent), and other topics. In the immediate post-earthquake environment, funding is often focused on lifeline facilities that sustained the most damage. Following the Whittier Narrows event, 35 percent of the lifeline funding went to dams, and 27.4 percent went to bridges and overpasses. The Loma Prieta Initiative allocated 39.4 percent of lifeline monies to water studies, 30.7 percent to transportation topics, 15.6 percent to power and 14.3 percent to gas. Notably absent in the Loma Prieta studies were studies of port or harbor facilities and telecommunications. It is interesting to note the inclusion of a significant number of socioeconomic topics in the research funded by the NSF (14.6 percent). Recently, the possibility of regional events with far-reaching impacts has prompted the study of earthquake impacts on the community in terms of direct damage, utility losses, and higher order, regional economic losses. Such research has required a multidisciplinary approach and is reflected in the variety of socioeconomic topics included in the Loma Prieta Initiative. These topics include not only community response and macroeconomic effects but secondary effects, such as those on the housing and rental markets and on the earthquake insurance industry. Other Research Efforts While publicly funded post-earthquake research has done a great deal to improve the state of the art in earthquake engineering, additional proprietary

OCR for page 135
Practical Lessons from the Loma Prieta Earthquake: Report from a Symposium Sponsored by the Geotechnical Board and the Board on Natural Disasters of the National Research Council research whose results are not readily available has also been conducted. Some of these results, however, are gradually reaching the public domain through various vehicles including: the Electric Power Research Institute, which has sponsored extensive research on the damage and vulnerability of facilities relevant to the electric power and nuclear industries; several workshops that serve to exchange information have been held, including a ''Wide Area Disaster Preparedness Conference'' (1991), which addressed such issues as performance, restoration, mitigation, and preparedness; the National Communications System, which co-sponsored a workshop in 1991 with the NSF entitled "Modeling the Impact of Major Earthquakes on Communications Lifelines: Research Accomplishments and Needs," and a second workshop in 1992 entitled "Assessment of State-Of-The-Art Approaches to Communication Lifeline Modeling for Earthquake Disasters; the purpose of these workshops was to "review the state of the art in modeling the effects of major earthquakes on communications lifelines and to assess the technical feasibility of developing models if none existed"; the presenters at this workshop discussed research results on various themes including seismic testing, actual performance, predictive damage and outage models, and mitigation and preparedness measures (NCS/NSF, 1991, in press); and conference proceedings from various professional organizations, such as the American Society of Civil Engineers/Technical Council for Lifeline Earthquake Engineering (ASCE/TCLEE) and the Earthquake Engineering Research Institute (EERI). Lessons Learned In this section, lessons learned as a result of research conducted after the Loma Prieta earthquake are discussed. Although many lessons have been documented, only those that have had a major impact on the analysis or design of lifeline systems are summarized. In general, lessons learned after major earthquakes fall into one of three categories: lessons that identify previously unknown seismic vulnerabilities; lessons that substantiate or contradict prior understandings of seismic vulnerability and/or design; and lessons that identify new variables with regard to vulnerability assessment and/or seismic design. In all three cases, some level of research is necessary to quantify the significance of the lesson learned or finding uncovered. The funding of post-earthquake research has been based, in part, on the amount of damage observed in the earthquake and the impact that the earthquake

OCR for page 135
Practical Lessons from the Loma Prieta Earthquake: Report from a Symposium Sponsored by the Geotechnical Board and the Board on Natural Disasters of the National Research Council FIGURE 5-7 Earthquake countermeasures for fiber-optic cables (Yagi et al. 1992). The Japanese are using this concept in the design of their underground manholes. One of the major concerns in the design of these facilities is that local liquefaction may uplift or displace these units. If this happens, there is concern that the cables contained within these units would either break or stretch. Therefore, in order to prevent possible failure, the Japanese have designed a number of different installations that incorporate cable slack as a design parameter. Several of these installations are displayed in Figure 5-7. Damage to telecommunication facilities will generally affect only local communication; the national telecommunication network is robust enough that outages in one part of the country should not affect other regions. Several studies performed for the National Communications System verified that on a nationwide scale, earthquakes should have little or no effect on calls within other regions. Several scenarios were run to determine the number of facilities that would be affected in a large earthquake on the Hayward fault in northern California. Based on fairly conservative damage criteria, it was determined that in such an event, the capacity of the network would drop to a possible low of about 67 percent in California and between 92 and 98 percent across the entire United States. These figures do not include, however, reductions in capacity caused by overload on the system. TRANSFERABILITY OF RESULTS One measure of the benefit of a particular research effort is the transferability of its methods, conclusions, or results to other areas of the country or to other types of facilities or systems. The transferability of earthquake research is critical, since many seismically active areas of the United States have not experi-

OCR for page 135
Practical Lessons from the Loma Prieta Earthquake: Report from a Symposium Sponsored by the Geotechnical Board and the Board on Natural Disasters of the National Research Council enced the effects of a large earthquake. Notable areas without modern, large earthquakes include the New Madrid Seismic Zone; the Wasatch Fault Zone; the Charleston, South Carolina area; the northeastern United States; and the Pacific Northwest. Many of the procedures available for the seismic analysis or design of lifeline systems and components are based on methods originally developed for California lifelines. In general, there are many similarities between California utility systems and systems in other parts of the United States. For example, the design and construction of major natural gas transmission pipelines, particularly interstate lines, are very similar, partly because their operation is federally regulated, and because they are designed and constructed under the same design guidelines. This is also the case for major oil pipelines. Because of these similarities, it is logical to assume that earthquakes exhibiting the same effects (such as liquefaction) would cause similar types of damage. Possible exceptions include the traveling wave effects that may be present in large midwestern earthquakes. These earthquakes may cause significantly more damage to underground pipelines or to structures sensitive to long-period effects (e.g., long-span bridges) than west coast earthquakes. Some of the lessons learned from the Loma Prieta earthquake that are considered transferable are: Pipeline damage models for ground failure effects (e.g., liquefaction) should be applicable in other parts of the United States. The ground failure effects observed in northern California after the Loma Prieta earthquake can and have occurred in other parts of the United States. Extensive liquefaction ground failures were observed during the 1811 and 1812 New Madrid earthquakes. It is likely that these types of ground failures will be responsible for the majority of the damage incurred by well-designed pipelines. Therefore, pipeline damage models developed primarily from California data should be applicable to earthquakes in other parts of the United States. Perhaps the one area where there is a lack of data in pipeline response data is in characterizing the effects of wave propagation. Data from other parts of the world (e.g., the 1985 Mexico City earthquake) will be needed to quantify the effects from this phenomenon. Potential damage to nonstructural elements in water treatment facilities could be more severe in earthquakes outside of California. Many of the treatment facilities (storage tanks, sedimentation basins) that suffered significant damage during the Loma Prieta earthquake are considered to be long-period sensitive structures. Sloshing effects were the primary cause of damage to these facilities. In a large New Madrid event, it is expected that larger areas will be impacted by ground motions containing low-frequency energy. As a result, facilities that are sensitive to these types of frequencies would be ex-

OCR for page 135
Practical Lessons from the Loma Prieta Earthquake: Report from a Symposium Sponsored by the Geotechnical Board and the Board on Natural Disasters of the National Research Council pected to experience similar or more extensive damage. Damage may also be more significant because of the longer durations expected for midwestern earthquakes. It is expected that after any significant U.S. earthquake, power outages in urban areas will last longer than in less developed areas, due to the need to inspect high-rise structures for gas leaks and ignition sources. The extended power outage in downtown San Francisco following the Loma Prieta earthquake resulted not from direct damage but from the need to perform building-by-building gas leak surveys prior to energizing the local power grid. While most of the city had power restored within a day of the earthquake, the high-rise district was without power twice as long—roughly 48 hours. Similar occurrences are anticipated for other U.S. earthquakes, and the impact could be far greater, particularly for a New Madrid type of event, which has the potential to simultaneously affect numerous highly developed urban areas. The use of cable slack in fiber-optic cable installations can reduce or eliminate potential interruption of service in any area expected to undergo significant movement or displacement. As a practical mitigation measure, this procedure should apply to any seismic region of the world. As indicated in previous discussions, this technique has been developed by NTT engineers to mitigate the effects of liquefaction on underground manhole structures. This technique may be particularly useful in the Midwest, where extensive liquefaction is expected in large earthquakes. FUTURE RESEARCH DIRECTIONS As a result of lessons learned from the Loma Prieta earthquake, several encouraging trends have developed. One of the more significant efforts is the collection and documentation of pipeline failure data. Only by collecting this information will researchers be able to validate analytical models for pipeline performance or to develop empirical models that apply over a wide range of seismic hazard effects and severity levels. Therefore, continued efforts to collect and document this-data are encouraged. Another area deserving further study is the area of ground failure assessment. It has been shown that pipeline performance is tied very closely with the types and levels of permanent ground displacement observed during an earthquake. If such a strong correlation exists, then the assessment of pipeline performance becomes a seismic hazard microzonation problem (i.e., identifying potential areas of ground failure and amounts of displacement). This further supports the continuation of efforts to map these types of hazards on a broad regional scale.

OCR for page 135
Practical Lessons from the Loma Prieta Earthquake: Report from a Symposium Sponsored by the Geotechnical Board and the Board on Natural Disasters of the National Research Council In earlier discussions of lessons learned, it was stated that system vulnerability methods were useful in identifying areas of potential outage. Ang et al. (1992) discussed the benefit of using these methods to quantify electric power system vulnerability; O'Rourke et al. (1990) used these methods to validate the vulnerability of the AWSS in the city of San Francisco and the rapid loss of water after the Loma Prieta earthquake. Although these methods have been used in areas outside of California, their application has not been widespread. One possible direction of future research is to ensure the development of appropriate regional models for system vulnerability assessment. Models that may vary from region to region include seismic hazard models (strong ground shaking, liquefaction, surface fault rupture, landslide, tsunami, and seiche) and seismic vulnerability or fragility models. The following recommendations for collaborative research are made: Stronger collaboration is needed between those researchers that characterize the severity of ground motions and ground failure effects (e.g., liquefaction) and those that model the seismic vulnerability of systems. Because lifeline systems cover large geographical areas, an assessment of seismic hazards on a broad regional scale is necessary. It is important that the appropriate seismic hazard measures are quantified and that this information is provided to the system modelers in the most useful format possible. Researchers who model the performance of lifeline systems must also coordinate their studies with social scientists. It is becoming clear that to describe the full impact of these catastrophic events, it is necessary to investigate secondary and higher order effects of the event. Previous studies (ATC, 1991) have stated that the more significant losses associated with the failure of lifeline systems will come from lifeline disruption and not from repair costs. Furthermore, social disruption costs, although difficult to quantify, may also be significant. Therefore, collaboration between engineers and social scientists must be strengthened. Stronger partnerships between the research community and industry must be developed. Many research efforts have benefited from such partnerships, and the results of these collaborations are more likely to lead to implementation of a study's recommendations. Currently, a federal initiative (spearheaded by the Federal Emergency Management Agency and the National Institute of Standards and Technology) to develop a plan for developing and adopting seismic design standards for public and private lifelines is underway. In order to ensure that such an effort is successful, it is essential that stronger partnerships between the research community, government agencies, and industry be developed. One of the most effective ways to initiate this partnership is to involve the end users (i.e., the lifeline operators) in government-sponsored research programs.

OCR for page 135
Practical Lessons from the Loma Prieta Earthquake: Report from a Symposium Sponsored by the Geotechnical Board and the Board on Natural Disasters of the National Research Council REFERENCES Ang, A., H-S.J. Pires, R. Schinzinger, R. Villaverde, and I. Yoshida. 1992. Seismic Reliability of Electric Power Transmission Systems—Applications to the 1989 Loma Prieta Earthquake. University of California at Irvine, prepared for the National Science Foundation and the National Center for Earthquake Engineering Research. ATC, 1991. Seismic Vulnerability and Impact of Disruption of Lifelines in the Conterminous United States. Applied Technology Council Report No. ATC-25, Redwood City, California. ASCE/TCLEE. 1992. ''TCLEE Pipeline Failure Database.'' Prepared for the National Science Foundation. Boheim, K.B., and C.M. Kelly. 1992. "Post-Earthquake Performance of Telecommunications Networks." Proceedings of the Fifth U.S.-Japan Workshop on Earthquake Disaster Prevention for Lifeline Systems, Tsukuba, Japan. EERI. 1990. Earthquake Spectra: Loma Prieta Earthquake Reconnaissance Report, Earthquake Engineering Research Institute, Supplement to Volume 6, May. Electric Power Research Institute. 1991. Proceedings: Wide-Area Disaster Preparedness Conference. EL-7298, EPRI, Palo Alto, California. EQE. 1990. The October 17, 1989 Loma Prieta Earthquake: Effects on Selected Power and Industrial Facilities. Prepared for the Electric Power Research Institute. Honegger, D.G. 1991. "Gas System Repair Patterns in San Francisco Resulting From the 1989 Loma Prieta Earthquake." In Proceedings of the Third U.S. Conference on Lifeline Earthquake Engineering, Monograph No. 4. Technical Council on Lifeline Earthquake Engineering, American Society of Civil Engineers. Isenberg, J., E. Richardson, H. Kameda, and M. Sugito. 1991. "Pipeline Response to Loma Prieta Earthquake." J. of Structural Engrg., 117(7), ASCE, New York, N.Y. Jones, B. 1993. "New Directions in Research, Societal and Economic Studies." Presented at the 1993 Annual Meeting of the Earthquake Engineering Research Institute, Seattle. Washington. Kennedy/Jenks/Chilton. 1990. 1989 Loma Prieta Earthquake Damage Evaluation of Water and Wastewater Treatment Facility Nonstructural Tank Elements . Prepared for the National Science Foundation, K/J/C 896086.00. Matsuda, E. 1993. Personal communication. Mohammadi, J., S. Alyasin, and D.N. Bak. 1992. "Investigation of Cause and Effects of Fires Following the Loma Prieta Earthquake." Illinois Institute of Technology, Report IIT-CE-92-01, NSF Grant BCS-9003557. NCS/NSF. 1991. Earthquake Workshop Proceedings: Modeling the Impact of Major Earthquakes on Communication Lifelines. Co-Sponsored by the National Communications System and the National Science Foundation. NCS/NSF. In press. Earthquake Workshop Proceedings: Assessment of State-of-the-Art Approaches to Communication Lifeline Modeling for Earthquake Disasters. Co-Sponsored by the National Communications System and the National Science Foundation. O'Rourke, T.D., H.E. Stewart, F.T. Blackburn, and T.S. Dickerman. 1990. Geotechnical and Lifeline Aspects of the October 17, 1989 Loma Prieta Earthquake in San Francisco. Technical Report NCEER-90-0001, NCEER, Buffalo, N.Y. O'Rourke, T.D., J.W. Pease, and H.E. Stewart. 1992. "Lifeline Performance and Ground Deformation During the Earthquake." The Loma Prieta, California Earthquake of October 17, 1989-Marina District. U.S. Geological Survey Professional Paper 1551-F, Washington D.C. PG&E. 1990. PG&E and the Earthquake of '89, Pacific Gas and Electric Company. San Francisco, California. Phillips, S.H., and J.K. Virostek. 1990. Natural Gas Disaster Planning and Recovery: The Loma Prieta Earthquake. Pacific Gas and Electric Company, San Francisco, California.

OCR for page 135
Practical Lessons from the Loma Prieta Earthquake: Report from a Symposium Sponsored by the Geotechnical Board and the Board on Natural Disasters of the National Research Council Seligson, H.A., R.T. Eguchi, L. Lund, and C.E. Taylor. 1991. Survey of 15 Utility Agencies Serving the Areas Affected by the 1971 San Fernando and the 1987 Whittier Narrows Earthquakes. Prepared for the Natural Science Foundation. Tang, A. 1992. "Technology Exchange with NTT on Seismic Protection of Telecommunication Facilities." Prepared for the National Science Foundation, ASCE/TCLEE. URS. 1988. "Risks of Earthquake-Induced Gas Fires in Residential Housing." Report prepared for Southern California Gas Company. Yagi, K., S. Mataki, and T. Sakurada. 1992. "Aseismic Countermeasure for Optical Fiber Cable in Liquefiable Ground." Proceedings of the Fifth U.S.-Japan Workshop on Earthquake Disaster Prevention for Lifeline Systems, Tsukuba Japan.

OCR for page 135
Practical Lessons from the Loma Prieta Earthquake: Report from a Symposium Sponsored by the Geotechnical Board and the Board on Natural Disasters of the National Research Council DISCUSSANTS' COMMENTS: LIFELINES Thomas D. O'rourke, Cornell University It is a pleasure to be here. I have four points I would like to make, some of which will echo those made by others. Finally, I would like to give a warning about our lifeline and infrastructure systems. Liquefaction-induced ground deformation was of key importance to the performance of the water supply in San Francisco and portions of the East Bay during the Loma Prieta earthquake. Correlations among areas of soil liquefaction and locations of buried pipeline damage show a clear pattern of system performance that depends on the severity of liquefaction and the spatial distribution of ground movement. These observations provide a practical framework for assessing the most vulnerable portions of the piping system and anticipating the effects of future earthquakes. These spatial observations can be used to come up with some simple rules useful for planning, emergency response, and development. Since buried systems depend so much on the deformation of the ground, the fate of the ground should be looked at as being, in part, the fate of these systems. Subsurface data have been used to characterize geometry and in situ properties of loose fills in areas of San Francisco, which were subject to liquefaction and ground failure. It has been found that mapping thickness of the submerged fills, a very simple parameter, is a good indicator of the severity of damage in a given area, particularly the damage to buried lifelines. The thickness of liquefaction fill or natural sand deposit is easily used in geographical information systems, providing an excellent vehicle for assessing urban hazards, microzoning for seismic hazard reduction, and planning for optimal lifeline performance during an earthquake. Computer simulations of damaged water supply performance during the earthquake are consistent with observations in the field and indicate that graphic modeling of hydraulic networks is sufficiently advanced for effective use in system management and emergency preparations. The computer simulations emphasize the importance of an independent power supply for isolation valves and the substantial effect that hydrant breaks have on water lost from the system. The events of the earthquake show that flexibility provided in San Francisco by the Portable Water Supply System was of critical importance in controlling and suppressing the fire that erupted in the Marina District. The ability to operate with portable hosing and draft from a variety of water sources, including underground cisterns and fireboats, provided a valuable extra dimension in the city's emergency response. Finally, I'd like to give a warning: beware the revenge of the infrastructure. We don't have to wait for an earthquake to have a major disaster. I hope the many valuable lessons learned after disasters can be applied to more effectively use our utility supplies and critical resources. Thank you.

OCR for page 135
Practical Lessons from the Loma Prieta Earthquake: Report from a Symposium Sponsored by the Geotechnical Board and the Board on Natural Disasters of the National Research Council Donald Ballantyne, Kennedy/Jenks Consultants Coming from Seattle, I want to talk about what effect the Loma Prieta earthquake has had in other areas of the country. I have comments on increasing earthquake awareness, water system evaluation and design, and emergency planning. Millions saw live TV coverage of the Loma Prieta event, which served to increase earthquake awareness. The water and sewer industry in particular had its awareness raised. The Water Pollution Control Federation was having its national conference in San Francisco that week, with many lifeline system owners from across the United States attending. They returned home to significantly influence the implementation of earthquake-mitigation programs in water and wastewater facilities. Moreover, many lifeline system owners sent teams to San Francisco to discuss the impacts of the event with their local counterparts. I believe that the closer one gets to an earthquake, the greater psychological impact it has. This can ultimately turn into the driving force to initiate earthquake-mitigation programs. The NEHRP program identified Seattle as a target area in 1987. In 1986, there were no lifeline earthquake mitigation programs in the Seattle area. In 1993, every major water and wastewater facility has a program in place, as do many of the moderate-size utilities. This resulted from the synergy of the NEHRP program's provision to develop basic seismological data in the northwest and the focus drawn from the Loma Prieta earthquake, which demonstrated what the effects of an event could be. Pipeline failures are concentrated in areas where liquefaction-induced permanent ground deformation occurs and can result in draining water storage tanks holding water needed for fire suppression. The old cast-iron pipe along the San Lorenzo River in Santa Cruz failed, which drained reservoirs. Service was lost to the city's two hospitals, and fire-protection capabilities were lost in those pressure zones. Luckily, fire was not a problem as there was no wind that evening. Power was not available for pumping to refill tanks for a number of days. This refocuses thinking on the hazards of liquefaction and ground deformation to vulnerable pipeline systems. System-control measures to maintain system function, and possibly dedicated fire-protection systems, are potential mitigation alternatives. With such a large inventory of pipe in the ground and the high cost of replacement, a more reasonable approach may be isolating damaged portions of the system—either the reservoirs themselves, pipelines crossing faults, or areas that are geotechnically unstable. Emergency planning and plan exercise is crucial. Historically, emergency planning has had a low priority. The Loma Prieta event demonstrated the need. All responsible organizations should be involved in emergency planning for their particular service. A few points particularly relevant to the water industry in-

OCR for page 135
Practical Lessons from the Loma Prieta Earthquake: Report from a Symposium Sponsored by the Geotechnical Board and the Board on Natural Disasters of the National Research Council clude provision for emergency power, pumps, chlorinators, and repair materials. Statewide or regional mutual aid programs are very worthwhile and should be improved. Thank you. Charles R. Roberts, Executive Director, Port Of Oakland Good morning. I am going to discuss this from the angle of implementing repair activities and some administrative issues that followed the earthquake. There is not much information available about how to repair the damage, how to get a facility back into operation quickly, or how to repair so that a facility won't fail in the next event. We had developed a reporting system that was activated automatically with a 5.0 event on the Hayward fault. In the event of an earthquake, supervisors informed the civil, mechanical, and electrical engineers to inspect and report back. We knew within hours where we needed to do emergency work and where we were out of business. The next step is to have a plan to assemble a work force, establish control centers, and initiate a multichannel communications system and a pre-authorized chain of authority. The system developed by the Port of Oakland was satisfactory, except more understanding of the chain of authority needed to be emphasized. Finally, time reporting systems, damage categories, and financial recording in accordance with and parallel to Federal Emergency Management Agency's procedures and policies must be set up ahead of time. For implementation, we now need to take these lessons learned to explain how to repair the subsidence problems so that they will not happen again and how to repair major concrete structures standing on long piles with heavy weights. Thank you. Steve Phillips, Pg&E Good morning. The most important item to stress is the need to be prepared. In 1987, PG&E developed a corporate emergency-operations-center concept designed to deal with system-wide emergencies such as an earthquake. We did go through a mock emergency exercise—using a scenario of a 7.0 earthquake on the Hayward fault. There have been other programs—in 1984 a formal gas pipeline replacement program was developed to look at several categories of pipeline that needed to be replaced on a systematic basis (cast-iron, pre-1930 steel-distribution facilities and certain types of older transmission lines with sub-standard welds).

OCR for page 135
Practical Lessons from the Loma Prieta Earthquake: Report from a Symposium Sponsored by the Geotechnical Board and the Board on Natural Disasters of the National Research Council Although begun in 1984 and funded at $80 million, this was a 20 to 30 year program. Almost all of the leakage occurring in the PG&E system was on facilities that fell into the pipeline replacement category. Distribution systems in the Marina District, Los Gatos, and Watsonville had to be replaced. Since the Loma Prieta earthquake, a seismic risk factor (soils and proximity to fault zones) has been included into the formula for prioritizing the pipe replacement schedule. In the mid-1980s, PG&E began to do seismic analysis of gas facilities using a pipeline scenario to see how they would fare in an event. The study had not been completed by 1989. There was no damage in those facilities—primarily because of the location and duration of the earthquake—, and PG&E was able to make the necessary modifications (which were fairly inexpensive). On the electric side, there was a similar program also begun in the mid-1980s. The Loma Prieta earthquake validated most of the assumptions made in that study. For example, there was no substation damage at the 115-kV and below level; there was the most damage at 500-kV substations; and there was substantial but less damage at 230-kV substations. At PG&E, we also predicted we would have no damage to control room facilities. That is exactly what happened during the Loma Prieta earthquake. A prioritized replacement program was already in place for most of the targeted equipment in those facilities. Although PG&E was not completely ready when the earthquake hit, we were positioned to be able to move forward rapidly to reduce the seismic risk when the situation did occur. Thank you.

OCR for page 135
Practical Lessons from the Loma Prieta Earthquake: Report from a Symposium Sponsored by the Geotechnical Board and the Board on Natural Disasters of the National Research Council This page in the original is blank.