6
Benefits from Performance-Based Engineering

Earthquake engineering has made significant advances in the past century. What began as an effort to protect lives from future earthquakes has grown to become an effort not only to protect life, but also to minimize damage and functional disruption to levels considered acceptable by owners and the communities in which their buildings are located. Earthquake engineers use “performance-based engineering” procedures to design structures with predictable and defined seismic performance. These procedures have been developed collectively, based on observations of the effects of major earthquakes worldwide. Over the past 50 years, these observations have been aided significantly by the availability of seismic monitoring data—records of earthquake events recorded by weak and strong motion instruments. All components of the built environment—buildings, bridges, roads, utility networks, and dams—share in the benefits of seismic monitoring, even though significant differences exist in how earthquake engineering is approached for each one.

The guidelines, standards, and codes available to earthquake engineers for the design of new structures and the rehabilitation of existing structures hold promise for protecting lives and the built environment against the largest expected earthquakes. Unfortunately, most communities in the United States that are threatened by such earthquakes are doing little to control their seismic risk. This may be partially because such communities—particularly in areas that have not felt shaking in historical time—are often skeptical about the scientific basis of earthquake forecasts and consider that the up-front cost of mitigating the risk is too high (EERI,



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 116
Improved Seismic Monitoring Improved Decision-Making: Assessing the Value of Reduced Uncertainty 6 Benefits from Performance-Based Engineering Earthquake engineering has made significant advances in the past century. What began as an effort to protect lives from future earthquakes has grown to become an effort not only to protect life, but also to minimize damage and functional disruption to levels considered acceptable by owners and the communities in which their buildings are located. Earthquake engineers use “performance-based engineering” procedures to design structures with predictable and defined seismic performance. These procedures have been developed collectively, based on observations of the effects of major earthquakes worldwide. Over the past 50 years, these observations have been aided significantly by the availability of seismic monitoring data—records of earthquake events recorded by weak and strong motion instruments. All components of the built environment—buildings, bridges, roads, utility networks, and dams—share in the benefits of seismic monitoring, even though significant differences exist in how earthquake engineering is approached for each one. The guidelines, standards, and codes available to earthquake engineers for the design of new structures and the rehabilitation of existing structures hold promise for protecting lives and the built environment against the largest expected earthquakes. Unfortunately, most communities in the United States that are threatened by such earthquakes are doing little to control their seismic risk. This may be partially because such communities—particularly in areas that have not felt shaking in historical time—are often skeptical about the scientific basis of earthquake forecasts and consider that the up-front cost of mitigating the risk is too high (EERI,

OCR for page 116
Improved Seismic Monitoring Improved Decision-Making: Assessing the Value of Reduced Uncertainty 2003). The combination of these two circumstances leaves the country in a state of increasing seismic risk. Although the hazard remains the same, the rapid expansion of the built environment nationwide—built without proper regard for earthquake potential—is causing the risk to grow steadily. With the exception of the West Coast, most states with seismic vulnerability do not have adequate building codes or policies requiring seismic design; no state in the nation has an adequate program for mitigating the expected unacceptable performance of existing buildings. To significantly diminish the growth of seismic risk, radical advances are needed that provide a refined understanding of earthquake hazards, and new analysis and design techniques are needed that more accurately accommodate expected ground motions. Current assessment and design procedures are based on simulation studies, laboratory tests, and post-earthquake field observations that result in generalized and conservative procedures for controlling damage. In recent earthquakes, comparison of building damage with ground motion recordings indicates that buildings have generally performed better than anticipated (Heinz and Poland, 2001). Accordingly, it is reasonable to expect that new techniques can be developed that will reduce seismic design requirements and thereby reduce the cost of seismic safety to more affordable levels. Seismic monitoring records hold the key to understanding how the built environment responds to damaging earthquakes and how best to fine-tune the design process so that the need is adequately—but not excessively—met. Monitoring alone is not sufficient to achieve this goal, but it is certainly a necessary component. The relatively modest funding required for significantly improving seismic monitoring and the subsequent development of new seismic mitigation techniques should be viewed in light of the potential for reducing the cost of constructing new facilities, strengthening existing structures to achieve proper performance, and avoiding losses after major damaging events. The roughly $200 million investment required for improved seismic monitoring and the cost of continuing research using the records should be viewed in light of the more than $800 billion invested annually in construction, the $17.5 trillion value of the built environment in the United States (FEMA, 2004), and the expected $100 billion plus loss from a single, major earthquake in an urban environment (EERI, 2003). SEISMIC MONITORING AND THE DEVELOPMENT OF EARTHQUAKE ENGINEERING Earthquake engineering—the application of science and technology to the design, rehabilitation, and repair of the built environment—has developed over the past hundred years as property owners experienced

OCR for page 116
Improved Seismic Monitoring Improved Decision-Making: Assessing the Value of Reduced Uncertainty unacceptable levels of damage caused by ground shaking. The techniques used in earthquake engineering have developed in parallel with the development of seismic monitoring programs. Because the need for design and construction is a daily process, engineers must carry out their work based on their best judgment, adding conservatism to deal with uncertainty. They use material science, structural analysis, large-scale testing, and the observation of failures to extend their understanding and fuel their intuition, and it is their collective intuition that is the basis for the prescriptive building standards and codes in use today. The goals of earthquake engineering have also evolved over the past hundred years. The earliest earthquake engineering attempts were undertaken both to eliminate damage and to neutralize the concerns raised by developers, bankers, and insurance companies. After the 1906 San Francisco, the 1925 Santa Barbara, and the 1933 Long Beach earthquakes, mandatory building codes emerged as a tool for eliminating the damaging effects of ground shaking. Seismic monitoring focusing on structures began in earnest during the late 1920s, and has grown steadily ever since. After the 1952 Kern County earthquake, a more rational goal emerged that was oriented toward protecting lives and keeping damage to repairable levels for all but the largest earthquakes. This goal remains the underlying principle of earthquake engineering today, and the goal of protecting lives has taken on a very broad definition. Seismic monitoring programs that are designed to provide improved understanding of the actual effects of earthquakes on the built environment are less than 50 years old. This, of course, is only an instant in geologic time, and so far only a small number of useful data sets have been obtained worldwide. Although much has been learned from those data sets, provoking dramatic changes to the design process, few records exist for sites that have experienced catastrophic damage from a design-level1 earthquake. The most important information that seismic monitoring networks are expected to provide has yet to be recorded. Seismic monitoring has led to development of national seismic hazard maps that are used to design structures with appropriate strength and durability. Although engineers once suggested—after the 1906 San Francisco earthquake—that the entire nation shared a common threat from earthquakes, it is now understood that the seismic hazard varies dramati- 1   A design-level earthquake represents the strongest ground motion expected to occur with a specified exceedence probability during the life of a building. Recent earthquakes in Kobe, Japan, and Taiwan were design-level earthquakes that provided a variety of strong motion records that have been useful for understanding the performance of structures in the immediate vicinity of the instrument.

OCR for page 116
Improved Seismic Monitoring Improved Decision-Making: Assessing the Value of Reduced Uncertainty cally across the nation. For example, whereas all of California and much of Nevada were once considered to be areas of highest hazard (ICBO, 1973), a combination of monitoring and scientifically defensible risk assessment has permitted the hazard to be mapped according to multiple zones that vary from moderate to highest hazard. Only the thin zone paralleling California’s great faults is now believed to be an area of highest hazard. Strong motion recordings have allowed seismic design procedures to be based on measurable parameters. Spectral values have replaced peak ground acceleration estimates as the key indicator of the severity of the earthquake hazard. Before monitoring systems were designed and installed, engineers could only speculate about the intensity of shaking based on the resulting damage. Now, in-structure instrumentation yields records that permit a clearer understanding of structural response during strong shaking, much as the “black-box” in an aircraft provides key information about conditions just prior to a crash. Monitoring provides insights into the characteristics and strength of the shaking that has caused damage. The few available seismic monitoring records show that structures experience earthquakes in a highly dynamic and time dependent manner, with the resulting structural performance due to complex combinations of the loading history, material strength, and structural design. Earthquake engineering design techniques have improved after each damaging earthquake and resulted in increasingly more advanced seismic design standards. When an earthquake occurs and structures experience more damage than their owners and engineers judge acceptable, the community of engineers adjusts design standards to avoid a repeat occurrence. Unfortunately, these advances in design are limited by the quality and quantity of the seismic monitoring records collected. When there are no records, there is a tendency to apply the new techniques to all buildings in all seismic environments. When there are records, the changes often apply only to construction in areas that are expected to experience a similar level of shaking. For example, earthquake engineers developed new procedures for the design and construction of moment-resisting steel-frame buildings based on the unexpected damaged that occurred during the 1994 Northridge earthquake. Unfortunately, no records were available at the sites where significant damage to moment-resisting steel-frame buildings occurred, so the subsequent research and resulting recommendations had to be based entirely on estimates of ground shaking. The resulting recommendations apply to the design of steel buildings nationwide. While they represent an important step forward in the design process, it is likely that they are more conservative than needed, mainly because strong motion records were not available to calibrate the observed damage.

OCR for page 116
Improved Seismic Monitoring Improved Decision-Making: Assessing the Value of Reduced Uncertainty The 1971 San Fernando earthquake produced the first regional set of strong motion records related to structures. The combination of seismic monitoring records and observations of damage resulted in code provisions requiring stronger, more resilient buildings that would experience less damage; the design of “essential buildings” that would remain functional; and the development of techniques for dealing with the massive inventory of old buildings that were constructed without sufficient seismic resilience. The San Fernando earthquake also provided the first evidence that structures that were designed using the same techniques did not necessarily perform in the same manner. It appeared that there were major differences in the ground motion from block to block, and significant differences in building performance due to the materials used in construction, the size and shape of the buildings, and the quality of construction. In order to adequately initiate the development of the next generation of design procedures to achieve the goals of earthquake engineering, specific recorded information is needed for a variety of construction styles and geologic conditions. After the 1971 San Fernando earthquake, seismic monitoring programs at federal and state levels were expanded to begin to provide the needed information. By the time of the 1989 Loma Prieta and 1994 Northridge earthquakes, hundreds of instruments had been deployed in the western United States and hundreds of records were collected. These instruments were focused on collecting three fundamentally different data types—free-field ground motion, lifeline response, and building response. They were deployed primarily in the most seismically active regions (i.e., the West Coast), with the expectation that the results could be extrapolated to all forms of construction nationwide. In parallel, earthquake engineers developed a variety of approaches to seismic design and rehabilitation that included permitting old, nonconforming buildings to remain in use or, in some cases, to be rehabilitated to a minimum “life safety” level. New buildings with normal use occupancy were designed to be “life safe” and “repairable” in the event of a large earthquake, and essential buildings were designed to be capable of operating after a major earthquake. While these standards were most often applied in California, they were also applied—especially for federally owned buildings—in other seismic regions. These various approaches to earthquake engineering were the forerunners of performance-based engineering. IMPROVEMENTS IN SEISMIC MONITORING NEEDED TO SUPPORT PERFORMANCE-BASED ENGINEERING Performance-based engineering began its formal conceptualization for practicing engineers after the 1994 Northridge earthquake, when the

OCR for page 116
Improved Seismic Monitoring Improved Decision-Making: Assessing the Value of Reduced Uncertainty Structural Engineers Association of California (SEAOC), with a grant from the Federal Emergency Management Agency (FEMA), produced Vision 2000: Performance Based Seismic Engineering of Buildings: Interim Recommendations (SEAOC, 1995). These recommendations were developed in response to the concern that the $40 billion cost of the Northridge earthquake was too high and the damage too disruptive to local communities. It appeared that the traditional goals of earthquake engineering related to protecting life were not sufficient. At the heart of the three volumes of data and interim recommendations in SEAOC’s report is the call for design procedures that produce buildings capable of performing at any one of a variety of predictable levels. The process has made the expected performance of structures during earthquakes more visible and has provoked discussion about the requirements for acceptable performance. The instruments deployed prior to the Loma Prieta and Northridge earthquakes yielded numerous strong motion records, resulting in opportunities to begin correlating observed damage with strong motion recordings and structural analysis techniques. Unfortunately, in both earthquakes there were no records taken in the immediate vicinity of any sites of major damage, and there were no records from significantly damaged bridges or other critical facilities. Detailed surveys conducted after the 1994 Northridge earthquake in the immediate areas surrounding the strong motion instruments illustrated once again the widely varying degrees of damage: much more variation was observed than would be expected based on the design standards in use (ATC, 2001). Unfortunately, the expanded instrumentation programs developed based on the 1971 San Fernando experience again failed to yield sufficient information to develop a credible understanding of the variations in ground motion and the reasons buildings performed as they did. The vast majority of buildings performed better than expected, so any notion that these records formed a basis for nationwide analysis and design was lost. Had there been thousands of instruments and records in the areas of greatest damage, there is no doubt that additional new advances would have been made in earthquake engineering with the consequent potential for considerable economic benefits. Another example of the inadequacy of the current seismic monitoring programs resulted from the 2003 San Simeon earthquake. Because this was an area of only moderate seismicity, few ground motion instruments and a single building instrumentation package had been installed. Given the intensity of the shaking and the lack of damage, an excellent opportunity to gain additional understanding of the damaging potential of moderate earthquakes was lost—an understanding that could have been applied in other areas of moderate seismicity throughout the nation. Opportunities for advances in seismic mitigation programs and tech-

OCR for page 116
Improved Seismic Monitoring Improved Decision-Making: Assessing the Value of Reduced Uncertainty niques will continue to be lost until sufficient instruments are deployed nationwide to capture the key characteristics of every damaging event. The development of performance-based engineering has resulted in a discussion concerning the level of damage that is expected, and provided the opportunity for owners and communities to determine what level of damage they will accept. Unfortunately, the engineering techniques needed to design to specific performance levels are still being developed. Efficient and cost-effective design standards are needed, because most procedures in use today yield expensive solutions that serve as deterrents to action. Needed Improvements To achieve the benefits that improved seismic monitoring can potentially provide, the following six enhancements to the current monitoring program are needed: The addition of sufficient free-field instruments nationwide to develop an understanding of the relationship between the source characteristics of an earthquake and the strong motions that are produced. Instruments also have to be located to identify the effects of the geologic setting and the local site conditions on ground motions, with a special focus on the seismic zonation of urban areas to identify locations where unusually strong ground motions are expected to occur. The end result that is required is a set of characteristic waveforms that can be used in design, based on an appropriate probabilistic assessment. These will provide the best possible input for an efficient design or assessment process and permit the proof testing of buildings in regions that experience the design-level ground motion. The addition of sufficient monitoring to identify the ground motion characteristics that trigger liquefaction, lateral spreading, and landslides. This will also allow the risk to be calculated in a manner consistent with similar calculations for the design of structures. This is important because the mitigation of geologic hazards is often carried out from a deterministic perspective, without regard for the probability of occurrence. Substantial savings in foundation costs, as well as expanded opportunities for building on otherwise questionable sites, will result. The addition of sufficient urban monitoring (i.e., the ideal extension to the present Advanced National Seismic System [ANSS] proposal) so that every structure that is damaged in an earthquake has a reference record that is suitable for understanding its performance. This does not require an instrument in every building—rather, there should be an instrument sufficiently close to record the ground motion that was experi-

OCR for page 116
Improved Seismic Monitoring Improved Decision-Making: Assessing the Value of Reduced Uncertainty enced. This equates to at least one instrument in every zip code and one instrument on every active geologic structure (fault or fold) within that zip code. The resulting records would provide the minimum amount of consistent information needed to understand the earthquake motion and provide the opportunity to develop the statistical data necessary both to calibrate assessment techniques and to develop appropriate performance indicators. The urban monitoring instruments proposed for the ANSS will partially accomplish this goal. The addition of sufficient structural monitoring of enough buildings nationwide to fully document the performance of all common building types during an event in terms of the lateral forces resisted, displacements experienced, the location and demand on elements developing ductility, and foundation-soil interaction. Instrumentation must allow for a complete determination of the demand on all structural and nonstructural elements. The resulting records will, in time, provide the data needed for the development of new analysis techniques that fully capture the linear and nonlinear performance of the structure. Current techniques are unable to estimate the deterioration of structural elements under strong shaking and therefore often overstate the significance of damage. In order to minimize the cost of seismic design and rehabilitation, more accurate techniques for estimating damage are needed. Special emphasis should be placed on instrumenting publicly owned buildings, especially federal buildings, to ensure continuity of maintenance of the instruments, open access to information about structural design and construction history, timely access to the monitoring records and to the buildings themselves so that recorded shaking levels can be correlated to actual building damage, and avoidance of liability issues that may concern private building owners. The development and deployment of new methods for monitoring buildings to directly record inter-story drift2 demand at critical locations from both structural and nonstructural perspectives. Current building instrumentation packages record acceleration, and integrate the waveforms to determine velocity and displacement. There is considerable controversy surrounding the accuracy of the calculated displacements, especially when they are used to calculate inter-story drift. Directly measured inter-story drift is expected to provide the most reliable ability to assess damage potential. 2   Inter-story drift is the amount of horizontal movement that occurs between floors during earthquake shaking. For example, if the tenth floor of a building deflects 20 inches and the ninth floor deflects 18 inches at the same time, the inter-story drift between the ninth and tenth floors is 2 inches.

OCR for page 116
Improved Seismic Monitoring Improved Decision-Making: Assessing the Value of Reduced Uncertainty Lifeline systems include transportation, water, wastewater, electric power, telecommunications, and gas and liquid fuel systems. They must perform successfully as complete systems to ensure uninterrupted operation of essential services. The addition of sufficient monitoring of lifeline systems to fully capture the interdependence of the related structures (e.g., pumping plants) and the interconnecting components (e.g., piping) in their distributed environment is required. This will allow a full understanding of the source and impact of element failures in the system that will lead to more robust designs. CALCULATION OF BENEFITS PROVIDED BY PERFORMANCE-BASED ENGINEERING When assessing the value of improved seismic monitoring as it relates to performance-based engineering, three parameters must be considered. These include the value of the built environment within the United States, the rate of construction, and the annual expected loss from earthquakes. Current estimates suggest that there are $17.5 trillion worth of structures and that 80 percent of construction is residential.3 The value of structures in states within high and very high seismic zones is about $5.8 trillion (33 percent of the total) and, when all states prone to seismic damage are included this amount, increases to about $8.6 trillion (49 percent of the total). For the purposes of this study, it is reasonable to assume that the average value of an instrumented building is $5 million. Construction estimates for 2004 totaled about $500 billion dollars for buildings and $400 billion for lifelines, and the annual construction value is expected to exceed $1 trillion per year within the next 10 years (FMI Corporation, 2004). As noted above, total annualized earthquake losses throughout the United States are estimated to be about $5.6 billion per year for buildings and building-related costs (FEMA, 2001a), and a single, major urban earthquake is expected to cause losses of more than $100 billion (EERI, 2003). Other critical assumptions about the built environment relate to the cost of seismic design and the cost of seismic rehabilitation. In an effort to establish a potential “ballpark” estimate of the benefits of improved monitoring, it is useful to consider the cost of seismic mitigation in general. Only broad-brush estimates are needed to encompass the various design styles and performance-based engineering techniques. Recent experience in various design practices has suggested (anecdotally) that the cost of including seismic design can range from 1 to 10 percent of a project 3   Based on the national inventory and valuation models contained in HAZUS-MH, released in early 2004.

OCR for page 116
Improved Seismic Monitoring Improved Decision-Making: Assessing the Value of Reduced Uncertainty budget. For example, a building that costs $150 per square foot to build without seismic design features would cost an additional $2 to $15 per square foot (a total of $152-165 per square foot) with seismic design. The difference relates mostly to the performance level selected, the sophistication of the design team, and its willingness to incorporate seismic design in the conceptual framework of the project and work to minimize its impact. Similarly, design office experience suggests that the cost of seismic rehabilitation can range from 10 to 150 percent of the replacement cost of the structure, depending on the structure, its condition, the seismic performance objective selected, and whether the structure will be occupied during reconstruction. A good generalized average cost is 20 percent of the replacement value. Another important assumption relates to the number of existing buildings that need seismic strengthening. Of the 50 states and the District of Columbia, 42 have some degree of earthquake potential, and 18 are considered to have high or very high seismicity. Based on a variety of building inventories and extensive seismic rehabilitation experience, it is reasonable to assume that about 10 percent of existing buildings within the earthquake-prone areas of the United States need seismic strengthening. Finally, since the majority of buildings remain in use until destroyed by natural disasters or neglect, all cost savings were calculated under the assumption that all buildings would eventually experience a design-level earthquake. The total value was then translated to an annualized cost by multiplying the total cost by 0.04.4 Seismic monitoring programs in place today will continue to generate benefits from performance-based engineering to the extent that they capture and record damaging events. The proposed improvements to the monitoring program are considered in terms of incremental improvements in seismic monitoring capabilities. The first is the augmentation of the United States National Seismic Network (USNSN) seismic monitoring backbone that will be provided as a component of USArray. The second is the implementation of the initial phase of the ANSS program, and the final step would be to add sufficient seismic monitoring nationwide to ensure that every damaging earthquake that occurred would be recorded to the extent necessary to advance the engineering design standards as much as possible. If these proposed enhancements are not done, the existing networks will continue to deteriorate due to age and obsolete technology, and eventually little seismic monitoring will exist to capture data from future 4   An approximate annualized value is derived by multiplying the dollar value of the capital stock by a plausible long-term interest rate, estimated as 4 percent.

OCR for page 116
Improved Seismic Monitoring Improved Decision-Making: Assessing the Value of Reduced Uncertainty earthquakes. The current implementation of USArray should lead to a long-term improvement in the understanding of the seismic hazard nationwide, although it will not provide any additional information related to the performance of structures in damaging earthquakes or provide any immediate benefit to the hazard assessment of the nation used in structural design. The implementation of ANSS—required to maintain what is currently available and achieve some of the enhancements stated above—should generate at least six significant benefits discussed below: two that are short-term (1 and 2) and four that are intermediate- to long-term (3 to 6). The extent and timelines of achieving these benefits will depend on significant earthquakes occurring in areas that are instrumented, as well as the timing of program funding. 1. Proof testing of instrumented buildings: Buildings that are instrumented and experience damaging earthquakes will provide new insights into how to better design buildings to predictable performance levels. They also will be candidates for “proof testing,” in that their performance capability for the recorded event will serve as a benchmark for performance during other earthquakes. The December 22, 2003, San Simeon earthquake may have been close to the maximum likely earthquake for that area and provides an example of this proof-testing benefit. The U.S. Geological Survey (USGS) probabilistic seismic hazard maps define design-level events for that part of the central California coastal area in terms of peak ground acceleration, as well as for short-period and 1.0-second spectral accelerations. Templeton Hospital, located in the area of strongest shaking, was instrumented and recorded strong motions at about the maximum design level expected. The building experienced only slight damage and did not experience any disruption of function. Like all hospitals in California, this building is currently slated for strengthening to meet new and stringent state requirements at an estimated cost of $20 million. Because the building has been essentially proof-tested, and the records of this testing are available, it is likely that no seismic strengthening is needed. Accordingly, about $50,000 worth of instrumentation and about $50,000 worth of instrument maintenance over the past 20 years will likely yield a 200-times benefit. Currently, approximately 300 buildings are instrumented nationwide, and this number will increase to approximately 600 under the ANSS program.5 As many as 50 percent of the instrumented buildings that need 5   The original calculations in a prerelease draft of this section were based on instrumentation of 3,000 buildings nationwide by ANSS. Clarification of implementation plans provided by USGS indicates that approximately 300 buildings will be instrumented with multiple sensors, so this figure is used in the calculations that follow.

OCR for page 116
Improved Seismic Monitoring Improved Decision-Making: Assessing the Value of Reduced Uncertainty rehabilitation are expected to pass their proof test once a design-level earthquake occurs. Engineering experience indicates that this will translate into an annualized savings of $3 million under ANSS and $665 million if all critical and economically significant buildings at risk of earthquake damage are eventually instrumented. 2. Post-earthquake repair of instrumented buildings: Structural engineers responsible for evaluating the post-earthquake condition of a building that is instrumented will have the advantage of knowing what level of ground shaking caused the observed damage, and they can determine how the shaking compares to the event for which the building was designed as well as the event that it has to be repaired to resist. This information will generally lead to lower repair costs, because the adequacy of the existing building will be better understood along with its key vulnerabilities—the repair and rehabilitation efforts can be better focused to address actual deficiencies. Engineering experience indicates that repair cost savings—ranging from 5 to 20 percent—are expected to occur for 20 percent of the currently instrumented buildings and 30 percent of the buildings to be instrumented under ANSS. Based on building inventory estimates, this is expected to translate into an annualized saving of $2 million. 3. Improved seismic hazard maps: The decision to design for seismic conditions—or rehabilitate because of seismic conditions—depends first on an understanding of the hazards anticipated at a particular building site. Although detailed site-specific seismic hazard studies can be performed, the costs of such studies are too high for most building projects. Seismic hazard maps have been available for decades to allow for a less rigorous assessment of seismic risk. Scientifically defensible maps were produced by USGS in 1997 that—for the first time—used seismic monitoring data in conjunction with engineering-based parameters. The limited distribution of the seismic monitoring data on which these maps are based has meant that they can only be used accompanied by a number of assumptions that lead to conservative assessments. Even with these limitations, these maps have refined our understanding of the nationwide distribution of earthquake hazards. Improved seismic monitoring using adequate free-field instruments is a critical requirement for further refining these maps. In addition, there is a need to better understand the relationship between the particular source characteristics of an earthquake and the strong ground motions that are produced. Instruments have to be located to identify the effects of the geologic setting and local site conditions on ground motions, with a special focus on the seismic zonation of urban areas to identify locations at which unusual ground motions may occur. Improved seismic monitoring of weak and strong ground shaking will ultimately lead to improvements in the hazard maps used for design. The expected improvements will have a direct impact on the cost of construction and the level of

OCR for page 116
Improved Seismic Monitoring Improved Decision-Making: Assessing the Value of Reduced Uncertainty damage experienced in those states that include areas of high or very high seismicity. Engineering experience suggests that an additional 1 percent savings in construction cost could occur with implementation of each of the incremental seismic monitoring programs—USArray, a revitalization of the USNSN, and full implementation of ANSS—as they provide improved seismic hazard information. This would translate to an annualized savings of $49 million for each of the incremental programs. 4. Refined analysis techniques: The ability of engineers to predict the performance of buildings during earthquakes depends on their ability to model the behavior for a specific ground motion. The techniques currently in use are based on available mathematical formulations and material properties. For the most part, they do not benefit from the combination of damage observations and recorded motion. There is evidence of considerable uncertainty in current predictions—using the best available analysis techniques—of the way buildings experience damage. Improved monitoring of buildings can eliminate this uncertainty, once sufficient waveforms are recorded and analysis techniques developed. This will ultimately result in reduced construction costs for new designs and for the rehabilitation of existing buildings. To achieve this goal, it will be necessary for sufficient urban monitoring to be added so that every structure that is damaged in an earthquake has a reference record that is suitable for understanding its performance. In addition, sufficient structural monitoring has to be added to enough buildings nationwide to fully document the performance of all common building types in terms of the lateral forces resisted, displacements experienced, the location and demand on elements developing ductility, and foundation soil interaction. Instrumentation must allow for a complete determination of the demand on all structural and nonstructural elements. Neither the current programs nor the enhancements to be provided by USArray and the revitalized USNSN will add this monitoring—it will only occur after full implementation of the ANSS program. The most significant impact of refined analysis techniques will be felt in rehabilitation projects. Based on an expectation that the cost of rehabilitation will decrease on the average by 3 percent, and considering the 10 percent of the inventory that is expected to need strengthening, this translates to an annualized savings of $34 million dollars. 5. Improved procedures for new construction: Improved seismic monitoring will provide the records needed to calibrate the earthquake engineering process, remove conservatism as appropriate, and reduce the cost of construction. This portion of the potential savings is interconnected with the ongoing research and testing programs related to seismic mitigation. The benefit is expected to manifest itself as a reduction in the cost of seismic mitigation and a reduction in the loss expected, with the latter anticipated

OCR for page 116
Improved Seismic Monitoring Improved Decision-Making: Assessing the Value of Reduced Uncertainty to be most significant. It is reasonable to expect that if the ANSS program is fully implemented, the average cost for new construction will decrease by 1 percent for 30 percent of the buildings built in the very high and high seismic regions. This translates into an annualized saving of $20 million. 6. Improved procedures for rehabilitation of existing construction: From the decision to consider seismic hazards, to the process for identifying the seismic deficiencies in a building, to the actual techniques for rehabilitating buildings, improved seismic monitoring will enhance understanding, reduce conservatism as appropriate, and make seismic mitigation more affordable and acceptable. The current assessment techniques, when judged against the performance of buildings in large earthquakes, appear to substantially overpredict the expected damage when seismic monitoring information is available to quantify the intensity of shaking. This tendency toward overprediction, when applied to an inventory of existing buildings being considered for seismic rehabilitation, will also overpredict the number that have to be strengthened. Unfortunately, this often results in very high estimates for the cost of mitigation that result in no action. The same thing often happens when communities attempt to develop public policies aimed at mitigating their seismic risks. The Unreinforced Masonry Building Ordinance adopted by the City of Paso Robles, when taken in the context of how buildings performed in the 2003 San Simeon earthquake, provides a good example. As required by California law, the City of Paso Robles had inventoried its unreinforced masonry buildings (UMB) and set a time schedule for their rehabilitation that extended through 2017. Unfortunately, the earthquake occurred sooner, and one building collapse resulted in two deaths. Of the approximately 20,000 buildings in and around Paso Robles, all but a handful performed without significant damage, even though they probably experienced ground motions near or above the design level. Current procedures for evaluating the seismic strength of buildings in that region, however, would probably show that a number of these buildings were not strong enough and were candidates for retrofit. Seismic monitoring will lead to improved assessment that will, in turn, lead to an increased focus on the buildings that actually have to be rehabilitated, thereby bringing rehabilitation program estimates down to a more acceptable size and allowing for inventories of hazardous buildings to be scheduled for mitigation, prioritized by their risk. In the case of Paso Robles, it appears that the one collapsed building should have been designated for rehabilitation many years ago. It is reasonable to expect that the annualized savings related to improved rehabilitation techniques will be similar to that calculated for refined analysis—an additional $34 million once the ANSS program is in place.

OCR for page 116
Improved Seismic Monitoring Improved Decision-Making: Assessing the Value of Reduced Uncertainty TABLE 6.1 Summary of Potential Design and Construction Benefits from Improved Seismic Monitoring Benefit Buildings Affected Total Value Seismic Costa Rehabilitation Cost Saved Annual Savings Beneficiary Proof testing of instrumented buildingsa 300 added by ANSS $3 billion $150 million $75 million $3 million Building owner Post earthquake repairb 300 added by ANSS $3 billion $315 million $63 million $2 million Building owner, FEMA Improved seismic hazard mapsc All buildings in seismic zones $165 billion $4.9 billion   $49 million Building owner Refined analysis techniquesd 10% of existing inventory Annual $170 billion $34 billion $850 million $34 million Building owner, FEMA Improved new construction procedurese All buildings in seismic zones $165 billion     $20 million Building owner, FEMA Improved rehabilitation proceduresf 10% of existing inventory Annual $170 billion $34 billion $850 million $34 million Building owner, FEMA Total annualized savings         $142 million   aSeismic cost is the cost to add appropriate seismic strengthening to a building during repair, rehabilitation, or initial cons truction. b50 % proof-tested, saving is from eliminating the need to rehabilitate. c20 % less repair costs. d1 % reduction in seismic cost. e5 % reduction in seismic cost. f2 % reduction in seismic loss for 30 % of the buildings.

OCR for page 116
Improved Seismic Monitoring Improved Decision-Making: Assessing the Value of Reduced Uncertainty SUMMARY This discussion has been rooted in observation and experience in the design and construction environment. It has attempted to generalize the issues sufficiently to allow simple estimates to be made that subjectively quantify the value of improved seismic monitoring. The total annualized savings described above amount to more than $140 million per year after implementation of the ANSS (summarized, together with identification of the benefit recipients, in Table 6.1). These calculations were also performed for the situation where 3,000 buildings nationwide are instrumented, rather than the approximately 600 planned as part of the existing ANSS proposal. These calculations indicate that such an expanded instrumentation program would provide potential annualized savings of about $250 million.