Earthquake Spectra

In 2008, an earthquake-planning scenario document was released by the U.S. Geological Survey (USGS) and California Geological Survey that hypothesizes the occurrence and effects of a M_w7.8 earthquake on the southern San Andreas Fault. It was created by more than 300 scientists and engineers. Fault offsets reach 13 m and up to 8 m at lifeline crossings. Physics-based modeling was used to generate maps of shaking intensity, with peak ground velocities of 3 m/sec near the fault and exceeding 0.5 m/sec over 10,000 km². A custom HAZUS^®MH analysis and 18 special studies were performed to characterize the effects of the earthquake on the built environment. The scenario posits 1,800 deaths and 53,000 injuries requiring emergency room care. Approximately 1,600 fires are ignited, resulting in the destruction of 200 million square feet of the building stock, the equivalent of 133,000 single-family homes. Fire contributes $87 billion in property and business interruption loss, out of the total $191 billion in economic loss, with most of the rest coming from shake-related building and content damage ($46 billion) and business interruption loss from water outages ($24 billion). Emergency response activities are depicted in detail, in an innovative grid showing activities versus time, a new format introduced in this study.

The ShakeOut Scenario is probably the most widely known and used earthquake scenario created to date. Much of the credit for its widespread dissemination and application lies with scenario development criteria that focused on the needs and involvement of end users and with a suite of products that tailored communication of the results to varied end users, who ranged from emergency managers to the general public, from corporations to grassroots organizations. Products were most effective when they were highly visual, when they emphasized the findings of social scientists, and when they communicated the experience of living through the earthquake. This paper summarizes the development criteria and the products that made the ShakeOut Scenario so widely known and used, and it provides some suggestions for future improvements.

The ShakeOut Scenario is premised upon the detailed description of a hypothetical M_w 7.8 earthquake on the southern San Andreas Fault and the associated simulated ground motions. The main features of the scenario, such as its endpoints, magnitude, and gross slip distribution, were defined through expert opinion and incorporated information from many previous studies. Slip at smaller length scales, rupture speed, and rise time were constrained using empirical relationships and experience gained from previous strong-motion modeling. Using this rupture description and a 3-D model of the crust, broadband ground motions were computed over a large region of Southern California. The largest simulated peak ground acceleration (PGA) and peak ground velocity (PGV) generally range from 0.5 to 1.0 g and 100 to 250 cm/s, respectively, with the waveforms exhibiting strong directivity and basin effects. Use of a slip-predictable model results in a high static stress drop event and produces ground motions somewhat higher than median level predictions from NGA ground motion prediction equations (GMPEs).

We apply a probabilistic method to develop fault displacement hazard maps and profiles for the southern San Andreas Fault. Two slip models are applied: (1) scenario slip, defined by the ShakeOut rupture model, and (2) empirical slip, calculated using regression equations relating global slip to earthquake magnitude and distance along the fault. The hazard is assessed using a range of magnitudes defined by the Uniform California Earthquake Rupture Forecast and the ShakeOut. For hazard mapping we develop a methodology to partition displacement among multiple fault branches based on geological observations. Estimated displacement hazard extends a few kilometers wide in areas of multiple mapped fault branches and poor mapping accuracy. Scenario and empirical displacement hazard differs by a factor of two or three, particularly along the southernmost section of the San Andreas Fault. We recommend the empirical slip model with site-specific geological data to constrain uncertainties for engineering applications.

An earthquake scenario, based on a kinematic rupture model, has been prepared for a M_w 7.8 earthquake on the southern San Andreas Fault. The rupture distribution, in the context of other historic large earthquakes, is judged reasonable for the purposes of this scenario. This model is used as the basis for generating a surface rupture map and for assessing potential direct impacts on lifelines and other infrastructure. Modeling the surface rupture involves identifying fault traces on which to place the rupture, assigning slip values to the fault traces, and characterizing the specific displacements that would occur to each lifeline impacted by the rupture. Different approaches were required to address variable slip distribution in response to a variety of fault patterns. Our results, involving judgment and experience, represent one plausible outcome and are not predictive because of the variable nature of surface rupture.

We compare simulated motions for a M_w 7.8 rupture scenario on the San Andreas Fault known as the ShakeOut event, two permutations with different hypocenter locations, and a M_w 7.15 Puente Hills blind thrust scenario, to median and dispersion predictions from empirical NGA ground motion prediction equations. We find the simulated motions attenuate faster with distance than is predicted by the NGA models for periods less than about 5.0 s After removing this distance attenuation bias, the average residuals of the simulated events (i.e., event terms) are generally within the scatter of empirical event terms, although the ShakeOut simulation appears to be a high static stress drop event. The intra-event dispersion in the simulations is lower than NGA values at short periods and abruptly increases at 1.0 s due to different simulation procedures at short and long periods. The simulated motions have a depth-dependent basin response similar to the NGA models, and also show complex effects in which stronger basin response occurs when the fault rupture transmits energy into a basin at low angle, which is not predicted by the NGA models. Rupture directivity effects are found to scale with the isochrone parameter.

The M7.8 San Andreas earthquake scenario for the ShakeOut exercise subjects more than a million wood-framed buildings to loads beyond their elastic capacity. Residential construction from the boom from the 1960's to 1980's relied heavily upon drywall sheathing and stucco for shear walls – more vulnerable than plywood or the gypsum lath and plaster of older buildings. During this same construction boom, many apartment buildings were built with tuck-under parking, and heavy masonry chimneys were prevalent. Based on HAZUS^®MH modeling we describe, more than 30,000 (mostly older) wood buildings could be red-tagged or yellow-tagged in the scenario event. More recent wood-frames, engineered using plywood shear walls, should perform well, even under the conditions produced by the San Andreas event considered. Cost-effective retrofit measures exist for some of the weaknesses found in older wood construction, but seismic upgrade of wood-framed buildings with structural wood panels remains expensive and intrusive.

This work represents an effort to develop one plausible realization of the effects of the scenario event on tall steel moment-frame buildings. We have used the simulated ground motions with three-dimensional nonlinear finite element models of three buildings in the 20-story class to simulate structural responses at 784 analysis sites spaced at approximately 4 km throughout the San Fernando Valley, the San Gabriel Valley, and the Los Angeles Basin. Based on the simulation results and available information on the number and distribution of steel buildings, the recommended damage scenario for the ShakeOut drill was 5% of the estimated 150 steel moment-frame structures in the 10–30 story range collapsing, 10% red-tagged, 15% with damage serious enough to cause loss of life, and 20% with visible damage requiring building closure.

This study examines the impact of the ShakeOut earthquake on reinforced concrete (RC) frame structures in Southern California. The assessment uses synthetic ground motions and nonlinear dynamic analysis to evaluate 20 RC frame buildings hypothetically located at 735 sites throughout the region. Results show that older nonductile RC frame structures may collapse at 8% to 32% of the sites analyzed, especially in Palm Springs, Los Angeles, and San Bernardino. Modern code-conforming RC frame structures are predicted to collapse at fewer sites (1–11%), but modern midrise construction may be vulnerable in Los Angeles due to rupture directivity and basin effects. These seismic performance metrics can inform the development of policies for emergency response and for mitigating earthquake-induced collapse of existing RC frame buildings. The study further provides a prototype that can be used in developing future scenario studies that will benefit from ongoing research to improve building and seismological models.

Fire following earthquake (FFE) is a significant problem in California. Potential FFE were examined for the ShakeOut Scenario assuming a M_w 7.8 event on a morning in mid-November, with breezy (10 mph) low humidity conditions. FFE is a nonlinear process whose modeling does not have great precision – in many cases the only clear result is differentiation between a few small fires versus major conflagration. For the scenario, analysis indicates approximately 1,600 ignitions, with the central Los Angeles basin experiencing hundreds of large fires. Estimated loss is hundreds to perhaps a thousand lives, and approximately 200 million sq. ft. of residential and commercial building floor area, corresponding to a loss of perhaps as much as one hundred billion dollars virtually fully insured. Mitigation opportunities include construction of a seismically reliable regional saltwater pumping system to protect central Los Angeles, and automated gas shut-off devices in densely built areas.

The ShakeOut Scenario assessed earth-science impacts, physical damage, and socioeconomic impacts of a hypothetical M7.8 southern San Andreas Fault earthquake. Among many detailed studies were special studies of 12 lifelines, 7 of which were performed by panels of employees of the utilities at risk. Panels met for four hours. Panelists were presented with the scenario's earth science impacts and previously estimated damage to “upstream” lifelines. They then hypothesized a realistic outcome of the earthquake on damage and service restoration, identifying research needs and mitigation options. The panel process worked well: panelists were well qualified and seemed to fairly assess realistic earthquake impacts and restoration, probably more realistically than an outside consultant would have been able to do, thus improving the ShakeOut. Panelists gained insight into lifeline interaction, mutual-aid needs, communication capabilities, and backup supplies. Southern California Edison, for example, enhanced its planning and preparedness for a large Southern California earthquake.

Seismic response simulations of the Los Angeles water supply to a M_w 7.8 San Andreas Fault earthquake scenario are used to assess the regional aqueduct and water distribution system performance in Southern California. Aqueducts sustain significant damage, and restoration of water flow is estimated to take between 4 and 18 months. Local emergency water supplies are insufficient to match the duration of aqueduct repairs, requiring severe water rationing. System serviceability declines rapidly due to numerous pipe leaks, causing serious difficulties for firefighting. Water service restoration to all customers is projected to take several months, with restoration of pre-earthquake water demand requiring more than a year. Business interruptions from long-term water rationing affect the regional economy greater than previously anticipated. Results from this scenario show how critical it is for all water agencies to prepare for a large-magnitude San Andreas earthquake.

The approaches necessary for estimating earthquake effects on railroads are different for developing design criteria or post-earthquake response policies and for developing railroad damage scenarios. In developing design criteria or post-earthquake response policies, the probability of ground motions exceeding a particular level is a primary concern. Developing damage scenarios, on the other hand, involves describing hypothetical effects for assumed ground motions. The identification of potential problems is the greatest benefit of disaster scenario development to railroads. Developing the Great Southern California ShakeOut Scenario revealed areas in which advance planning and arrangements by the affected railroads could reduce delays in repair work or improve the efficiency of operation during recovery. These include arranging emergency waivers for permits and similar governmental requirements, developing arrangements to accommodate earthquake-related conflicts between commuter and freight operations, advance arrangements for emergency use of helicopters, and physically securing equipment at the San Bernardino dispatching center to reduce damage.

This paper describes a hypothetical scenario of public response to a large regional earthquake on the southern section of the San Andreas Fault. Conclusive social and behavioral science research over decades has established that the behavior of individuals in disaster is, on the whole, controlled, rational, and adaptive, despite popular misperceptions that people who experience a disaster are dependent upon and problematic for organized response agencies. We applied this knowledge to portray the response of people impacted by the earthquake focusing on actions they will take during and immediately following the cessation of the shaking including: immediate response, search and rescue, gaining situational awareness through information seeking, making decisions about evacuation and interacting with organized responders. Our most general conclusion is that the actions of ordinary people in this earthquake scenario comprised the bulk of the initial response effort, particularly in those areas isolated for lengthy periods of time following the earthquake.

An M_w 7.8 earthquake as described in the ShakeOut Scenario would cause significant damage to buildings and infrastructure. Over 6 million tons of newly mined aggregate would be used for emergency repairs and for reconstruction in the five years following the event. This aggregate would be applied mostly in the form of concrete for buildings and bridges, asphalt or concrete for pavement, and unbound gravel for applications such as base course that goes under highway pavement and backfilling for foundations and pipelines. There are over 450 aggregate, concrete, and asphalt plants in the affected area, some of which would be heavily damaged. Meeting the increased demand for construction materials would require readily available permitted reserves, functioning production facilities, a supply of cement and asphalt, a source of water, gas, and electricity, and a trained workforce. Prudent advance preparations would facilitate a timely emergency response and reconstruction following such an earthquake.

Recovery from an earthquake like the M7.8 ShakeOut Scenario will be a major endeavor taking many years to complete. Hundreds of Southern California municipalities will be affected; most lack recovery plans or previous disaster experience. To support recovery planning this paper 1) extends the regional ShakeOut Scenario analysis into the recovery period using a recovery model, 2) localizes analyses to identify longer-term impacts and issues in two communities, and 3) considers the regional context of local recovery. Key community insights about preparing for post-disaster recovery include the need to: geographically diversify city procurement; set earthquake mitigation priorities for critical infrastructure (e.g., airport), plan to replace mobile homes with earthquake safety measures, consider post-earthquake redevelopment opportunities ahead of time, and develop post-disaster recovery management and governance structures. This work also showed that communities with minor damages are still sensitive to regional infrastructure damages and their potential long-term impacts on community recovery. This highlights the importance of community and infrastructure resilience strategies as well.

For the ShakeOut Earthquake Scenario, we estimate $68 billion in direct and indirect business interruption (BI) and $11 billion in related costs in addition to the $113 billion in property damage in an eight-county Southern California region. The modeled conduits of shock to the economy are property damage and lifeline service outages that affect the economy's ability to produce. Property damage from fire is 50% greater than property damage from shaking because fire is more devastating. BI from water service disruption and fire each represent around one-third of total BI losses because of the long duration of service outage or long restoration and reconstruction periods. Total BI losses are 4.3% of annual gross output in the affected region, an impact far larger than most conventional economic recessions. These losses are still much lower than they potentially could be due to the resilience of the economy.

Following a damaging earthquake, “business interruption” (BI)—reduced production of goods and services—begins and continues long after the ground shaking stops. Economic resilience reduces BI losses by making the best use of the resources available at a given point in time (static resilience) or by speeding recovery through repair and reconstruction (dynamic resilience), in contrast to mitigation that prevents damage in the first place. Economic resilience is an important concept to incorporate into economic loss modeling and in recovery and contingency planning. Economic resilience framework includes the applicability of resilience strategies to production inputs and output, demand- and supply-side effects, inherent and adaptive abilities, and levels of the economy. We use our resilience framework to organize and share strategies that enhance economic resilience, identify overlooked resilience strategies, and present evidence and structure of resilience strategies for economic loss modelers. Numerous resilience strategies are compiled from stakeholder discussions about the ShakeOut Scenario (Jones et. al. 2008). Modeled results of ShakeOut BI sector losses reveal variable effectiveness of resilience strategies for lengthy disruptions caused by fire-damaged buildings and water service outages. Resilience is a complement to mitigation and may, in fact, have cost and all-hazards advantages.

The Great Southern California ShakeOut was a week of special events featuring the largest earthquake drill in United States history. On November 13, 2008, over 5 million Southern Californians pretended that the magnitude-7.8 ShakeOut scenario earthquake was occurring and practiced actions derived from results of the ShakeOut Scenario, to reduce the impact of a real, San Andreas Fault event. The communications campaign was based on four principles: 1) consistent messaging from multiple sources; 2) visual reinforcement: 3) encouragement of “milling”; and 4) focus on concrete actions. The goals of the ShakeOut established in Spring 2008 were: 1) to register 5 million people to participate in the drill; 2) to change the culture of earthquake preparedness in Southern California; and 3) to reduce earthquake losses in Southern California. Over 90% of the registrants surveyed the next year reported improvement in earthquake preparedness at their organization as a result of the ShakeOut.

This paper demonstrates an innovative approach for learning about earthquake victims’ behavioral responses to an emergency situation in the immediate aftermath of an earthquake. Researchers developed a scenario following a magnitude 7.8 earthquake that leads to escalating complications described over eight episodes. Subjects were assigned to scenario simulation groups (SSG) and instructed to discuss how they would cope with problems as if they were experiencing the scenario. Subjects first discussed their reactions and potential decisions they might have to make as a group. They were then asked to individually record their behavioral intentions, cognitive reactions (concern) and emotional state (fear) in a survey instrument. Subjects’ responses were tracked over the eight episodes of the scenario. The SSG methodology yielded a more realistic understanding of how a respondent's reactions and behavior change in the immediate aftermath of an earthquake. The implications of the SSG approach on disaster preparedness and response are discussed.