Earthquake Spectra

This paper is an abridgement of a classical mathematical solution of twelve of fifteen translational modes of vibration of a fifteen story building. This work was done just prior to the advent of the digital computer, by Edward C. Robison (C) in the period 1948-1950. He used seven place logarithms! The solution is in terms of lateral forces in each of twelve modes at each of fifteen floors coming out of admissible solutions for eigenvalues from the characteristic equation. These forces, when “normalized” by a constant multiplying factor for each mode, add algebraically to the weight of each floor, and in the over-all to the weight of the building—a principle illustrated by Robison's work and by Figure 9 of the Joint Committee's report (a) for which figure Milton Ludwig (D) was responsible, independently, about the same time.

Private procurement of earthquake insurance is often advocated as a preferred means to mitigate the economic consequences of severe earthquake damage. For recognized reasons, a large majority of residential property owners have demonstrated a marked reluctance to purchase earthquake insurance. Suggestions that the property damage/casualty insurance industry and/or the mortgage finance industry act as catalysts to stimulate widespread private acquisition of earthquake insurance protection have not been embraced by either industry for a number of understandable reasons. Nonetheless, interest in, and the potential for, such intervention remains strong, and many current obstacles might prove surmountable. One such obstacle is the risk of antitrust violations arising in the course of the interactions which might be necessary if either or both referenced industries actively engaged in creating a widespread demand for earthquake insurance. Legal counselling by antitrust specialists could permit such a risk to be largely avoided and substantially diminished.

A summary of the different provisions existing in Chile is given. The codes related to earthquake engineering practice are divided into two major groups: the first defining the loads and actions, and the other dealing with materials behavior, resistance, and detailing requirements. The overall characteristics of these codes are discussed. Special emphasis is given to the seismic design code provisions that define the level of earthquake action to expect, depending on building characteristics and site soil conditions. No seismic risk map is included in the code. Two different analysis procedures are allowed: an equivalent lateral forces procedure where torsion is considered through an amplification of the static torsion in the building; and the standard response spectrum analysis method with a three degree of freedom per story model of the building. The maximum responses of the different modes are combined using a special combination rule. Additional restrictions are imposed to torsional effects, and to overall building deformations. Finally, the basic ideas being discussed in the revision that is actually being done to the code are presented.

Seismic risk analysis involves determining the adverse consequences that people and society might suffer as a result of future earthquakes, and estimating the probability of these consequences for some future time period. We review the methods used, and present a simple example for a hypothetical building in Los Angeles. The purpose of a seismic risk analysis is to make informed decisions about seismic safety, and this is illustrated with the Los Angeles example by presenting the implications of several options available for a property owner to accept, insure, or mitigate seismic risk.

The 1985 experience in Mexico City has again demonstrated that building damage from earthquakes is related to the ground conditions beneath the building. When the natural period of the building matches the natural site period, resonance can occur, resulting in large building deformations and severe damage. Mapping dynamic site periods for a city, and governing building construction accordingly can greatly reduce the threat of damage from earthquakes. Using dynamic soil property data and a one-dimensional linear elastic computer program, a dynamic site period map for Charleston, SC, is developed and presented. Comparison of the building damage distribution for the 1886 earthquake with the distribution of dynamic site periods shows the important role of dynamic site period. An approximate method of estimating the linear site period is given.

Owing to Iraq's unique geographic location (29-38.5 N;39-49 E) the eastern and northeastern parts of the country are directly influenced by the seismic activity of the Tauros-Zagros tectonic zones. An earthquake data file containing nearly 550 entries covering the period 1905-1984 is compiled and an analysis of completeness is performed using Stepp's (1973) model. It is observed that the file is complete for 4.8 Ms and greater over the whole 80-year sample of data while during the past 25 years this threshold drops to about 4.1 Ms. The “completed” data file is then employed in re-evaluating the magnitude-frequency formula for the country. To further investigate the problem of earthquake occurrence a simple statistical treatment is applied to a “constant-b” seismicity model whereby an increase in the frequency of occurrence of approximately 35% is identified in the latter half of the seismic history of the country.

This review paper compares ANSI, NEHRP, SEAOC, and UBC. A few essential differences among these documents are as follows: (a) The NEHRP document gives force levels corresponding to a strength-based or limit states design, while the other three documents give force levels that correspond to working or service stress design; (b) the importance factor is used as a multiplier of base shear level in all documents except NEHRP, which treats building importance by a seismic hazard exposure group; (c) NEHRP and UBC-1988 contain detailing requirements for all common construction materials and all seismic zones, while UBC-1985 contains detailing requirements for zones of high seismicity but only limited requirements for zones of moderate seismicity; (d) P-delta analysis is specified by NEHRP for all buildings that must be analyzed, by SEAOC for buildings that exceed drift limits, by UBC-1988 for all buildings except those in Zones 3 and 4 meeting drift limits, and is not specified by ANSI.

The vulnerability of important high voltage switchyard equipment to low seismic input motions has again been demonstrated. Ten of sixteen 230 kV circuit breaker phases failed at the Edmonston Pumping Plant on the California Aqueduct. Failures were primarily porcelain members. The failures were due to both dynamic response of the members and to a lack of slack in conductors connecting adjacent equipment. The aqueduct was out of service for four days until power was partially restored. Repairing or replacing all damaged equipment is expected to take from 6 to 8 weeks. The importance and vulnerability of high voltage power equipment, the long time required to make repairs, and the difficulty in obtaining spare parts quickly emphasizes the need for cost effective measures to improve the seismic response of this type of equipment. While this earthquake did not substantiate the effectiveness of a base isolation system installed on one circuit breaker, it did illustrate the need for instrumentation to evaluate the performance of these systems. Leaks developed at two Smith-Blair mechanical couplings in one of two 192″ diameter aqueduct siphons near the pump plant. Whether this was due to ground deformations or inertial loads could not be determined.

A wedging system for coupling downhole accelerometers into cased boreholes has been developed and deployed at an EPRI/USGS instrument array located in the Cholame Valley near Parkfield, California. The wedging system allows downhole accelerometers to be set and retrieved by hand. With this system accelerometers can be readily retrieved from cased holes for inspection, maintenance, or redeployment.