SMIP 1998 Seminar

The principal objective of this study is to develop an improved parameterization for the engineering specification of near-fault ground motions. In addition to the response spectrum, we would like to include time domain parameters that describe the near-fault pulse, such as its period, amplitude, and number of half-cycles. We have developed a preliminary model that relates time domain parameters of the near-fault ground motion pulse to the earthquake magnitude and distance. The model is for forward rupture directivity conditions, which produce a strong pulse of motion in the fault-normal direction. The pulse parameters that we have modeled are the period and peak amplitude of the largest cycle of motion of the velocity pulse. The records analyzed include 15 time histories recorded in the distance range of 0 to 10 km from earthquakes in the magnitude range of 6.2 to 7.3, augmented by 12 simulated time histories that span the distance range of 3 to 10 km and the magnitude range of 6.5 to 7.5.

The results of the study summarized here are intended to shed light on some of the important issues that affect the response of structures to near fault ground motions. Attempts are presented to identify salient response characteristics, to describe near fault ground motions by properties of equivalent pulses, and to utilize the pulse response characteristics to define behavior attributes of structures when subjected to near fault ground motions. The ultimate objective is to develop improved design procedures. Preliminary recommendations are made in this respect, but it must be understood that much more work needs to be done before comprehensive answers will be found.

When a potentially damaging earthquake occurs, utilities (electricity, gas, water, and telecommunications) have an urgent need for information about the effects of the event so that they can make optimal decisions regarding safety and maintaining and restoring utility functionality. Modern earthquake instrumentation systems, including strong- motion recorders and regional seismic networks, can collect data and provide information products that can greatly improve this decision-making and action-taking process. Four areas of utility response to earthquakes illustrate the utilization of these data: (1) Rapidly available network and strong motion data can provide an earthquake alert that will make utility personnel aware that an earthquake is occurring, what area of the utility's service territory is affected, and the likely extent of damaging ground motions. This alert will focus the earthquake response attention of the utility and may permit quick operational and life-safety actions. (2) Within 10 to 30 minutes after the earthquake, analysis of strong-motion data from key utility sites will provide assessments of the likelihood of damage that can be used to prioritize deployment of field personnel and guide the initial operational control and recovery plans. (3) In the same time frame, similar strong-motion-based damage assessments of transportation routes (e. g. freeways, bridges, and overpasses) along with reported damage and disruption will help the utilities plan how to get inspection and repair crews to key facilities. In addition, damage likelihood assessments of commercial, industrial, and residential buildings will indicate where utility service connections may need rapid responses to safety and secondary damage threats. (4) Within a few hours of the earthquake, pre-arranged building inspectors can use building response strong-motion measurements to help evaluate the safety of continued occupancy of structures housing critical post-earthquake response functions.

As the number of large civil structures instrumented for strong motion is increasing, efforts towards utilizing the earthquake data collected from these structures is also increasing. The studies are geared towards verifying seismic engineering design assumptions by comparing the theoretical models to the actual readings. Efforts to utilize the data ranging from simple comparison of the estimated structural period of vibration with the recorded free vibration, to complex comparisons of non-linear time-history models are underway. Many more studies are needed to take full advantage of this valuable data.

Accurately monitoring bridge movements during a large earthquake is necessary to advance our understanding of how these massive structures are affected by seismic input. Bridges of different structure types react differently to the same seismic wave patterns. Dynamic soil-structure interaction can be studied and theories can be verified or disproved based on the actual readings. Before strong motion sensors were placed at ground sites or on civil structures, theories were based on very little data. Therefore, the data collected from large earthquakes with these sensors are invaluable to the seismic engineering community.

The California Department of Transportation (Caltrans) and the California Strong Motion Instrumentation Program (CSMIP) of the California Department of Conservation's Division of Mines and Geology have instrumented more than 50 Caltrans bridges throughout the State since the 1989 Loma Prieta earthquake. In addition, CSMIP and Caltrans are installing more near-real-time stations at selected bridge sites in the State. Consequently, more near-real-time strong- motion data will be available quickly after an earthquake. These data provide information on ground shaking and response of the bridge structure, and are useful not only for improving seismic design practices but for post-earthquake damage evaluation of bridges. This paper describes the current status and future plan of the Caltrans/CSMIP bridge instrumentation project, and discusses quick application of strong-motion data to post-earthquake evaluations of bridges. Cases of quick application of near- real-time data are presented and criteria for determining post-earthquake inspection of bridges are discussed.

The response of an instrumented reinforced concrete moment-resisting frame (RCMRF) building, located in Southern California, was investigated in this research and compared to the response of linear elastic analytical models of the building. RCMRF buildings are particularly difficult to model when the objective is to predict the performance of the building. Therefore, nine models of the case study building were created by making three assumptions for the stiffness of the beams and columns and three assumptions for the stiffness of the beam-column joints. Fundamental periods for the models were compared to the fundamental periods calculated directly from the building response recorded during the 1987 Whittier and 1994 Northridge earthquakes. In addition, the analytical models were subjected to the ground accelerations recorded during the Whittier and Northridge earthquakes in a time history analysis and the maximum floor displacements compared to the recorded floor displacements.

As a part of a project sponsored by the Strong Motion Instrumentation Program of state of California (SIMP), responses of three concrete frame buildings during the 1994 Northridge earthquake are being evaluated. This investigation is at various stages of progress for the three buildings. For the 13 Story Sherman Oaks Building both linear and nonlinear analysis have been performed and nonstructural components response has been evaluated. For the 32 Story Burbank Building, linear analysis had been completed and nonlinear analysis is in progress. For the 7 Story Van Nuys Hotel construction of computer models are in progress. This paper serves as a status report on what has been learned so far from these investigations. Further details and refinements to observation presented in this paper will be forthcoming in a report to the SMIP.

This paper describes the key features and components of the regional earthquake loss estimation methodology developed by the National Institute of Building Sciences (NIBS) with funding provided by the Federal Emergency Management Agency (FEMA). The FEMA/NIBS earthquake loss estimation methodology is intended primarily to assist emergency response planning and mitigation efforts of state, regional and local community governments. The FEMA/NIBS methodology incorporates state-of-the-art approaches for characterizing earthquake hazards, including ground shaking, liquefaction and land-sliding; estimating damage and losses to buildings and lifelines, estimating casualties, shelter needs and economic losses.

Of particular importance is the use of quantitative measures of ground shaking hazard (i.e., response spectra) in the estimation of building damage. Damage and loss are based on ground motion data, rather than on Modified Mercalli Intensity (MMI) commonly used by other earthquake loss estimation methods. While the FEMA/NIBS methodology was developed primarily for pre-earthquake planning purposes, the use of response spectra to predict damage makes the technology potentially valuable as a post-earthquake processor of near-real-time data from strong-motion networks such as the TriNet system. It is suggested that the FEMANIBS methodology be interfaced with strong-motion instrumentation networks to better assist post earthquake response and recovery efforts. It is also suggested that predictions of earthquake damage and loss be used to assist locating strong-motion instruments in areas where buildings and other infrastructure are most at risk.

Summarized in this paper are the results of two recent investigations utilizing recorded motions of buildings to develop improvements in two aspects of seismic code provisions for buildings: (1) fundamental vibration period formulas, and (2) accidental torsion.

Rapid (3-5 minutes) generation of maps of ground motion shaking and intensity is accomplished with advances in real-time seismographic data acquisition combined with newly-developed relationships between recorded ground motion parameters and expected shaking intensity values. Estimation of shaking over the entire regional extent of southern California is accomplished by spatial interpolation of the measured ground motions with geologically-based, frequency and amplitude-dependent site corrections. Production of the maps is automatic, triggered by any significant earthquake in southern California. Maps are now made available within several minutes of the earthquake for public and scientific consumption via the World-Wide-Web; they will be made available with dedicated communications for emergency response agencies and critical users.