Earthquake Ground Motion

A total of sixty earthquake ground motion records were developed by Somerville, et al. (1997) as part of the SAC Joint Venture1 research program. Suites of twenty ground motion records were developed for each of three seismic hazard categories having exceedance probabilities of 50% in 50 years, 10% in 50 years, and 2% in 50 years. These exceedance probabilities correspond to 72-yr, 475-yr, and 2475-yr return periods, respectively. The ground motion records for each seismic hazard category are listed in Tables C. 1 through C.3. For each ground motion record, these tables provide the ground motion identification label, e.g. LA01, the location of the recording site, the name of the earthquake, the earthquake magnitude, the distance from the earthquake source to the recording station, the total duration of the earthquake record, and the peak ground acceleration (PGA). The earthquake records were selected to represent the range of earthquake magnitudes and distances expected to contribute to the seismic hazard at a building site located in Los Angeles, California. The earthquake ground motion database includes both recorded (LA01-LA30; LA41-LA60) and simulated (LA31-LA40) time

1 SAC is a joint venture of the Structural Engineers Association of California (SEAOC), the Applied Technology Council (ATC), and California Universities for Research in Earthquake Engineering (CUREe).

histories. The ten simulated time histories were generated using the broadband simulation procedure described in Somerville, et al. (1997).

For each seismic hazard category, a target response spectrum for firm soil conditions was developed based on the 1996 USGS probabilistic ground motions (Frankel et al., 1996; 1997a; 1997b). The target response spectra, shown in Figure C.l, are defined at periods of 0.3, 1.0, 2.0 and 4.0 seconds for each seismic hazard category. The USGS probabilistic ground motions maps provide 5% damped response spectrum values at periods of 0.1 to 2.0 seconds for soft rock soil conditions throughout the United States. To obtain the target response spectrum values at a period of 4.0 seconds, a probabilistic seismic hazard analysis was performed by Abrahamson (Somerville, et al., 1997) at 0.3, 1.0, 2.0, and 4.0 seconds, and the results used to extrapolate the USGS target spectra. The target spectra were scaled further, as described in Somerville, et al. (1997), to account for the difference in soil conditions. In addition, the ground motion records were uniformly scaled in order to minimize the error between the average response spectrum for each pair of time histories and the target response spectrum. The scale factor associated with each record is given in Tables C.l through C.3. Scaling the records in this manner preserves the original shape of the response spectra and maintains the ratio between component spectral values for each pair of time histories.

Tables C.l through C.3 quantify the duration of strong motion for each earthquake record. The duration of strong motion, tD, follows the definition by Trifunac and Brady (1975), which is stated as follows,

where toos and t0J)S correspond to the times during the earthquake record at which the Arias intensity (Arias, 1970) reaches 5% and 95 %, respectively, of its total value. The total value of the Arias intensity, IA, is defined by the following, where T is the duration of the earthquake record and xg(t) is the time history of ground acceleration.

A 2% damped response spectrum analysis was performed, for each of the sixty earthquake ground motion records listed in Tables C. 1 through C.3 using the SPECTRA (Ewing, 1992) computer program. The damping was selected to correspond with the level of Rayleigh damping used in the nonlinear time history analyses of the three model steel frame buildings as discussed in Section B.2.5 of Appendix B. The response spectra are shown in Figures C.2 through C.7 for spectral acceleration values and Figures C.8 through C.13 for spectral displacement values. Figures C.14 through C.19 show the composite response spectra for the sixty earthquake ground motion records. As discussed in Section 3.5.2 of Chapter 3, a composite response spectra is a graph that shows the variation in spectral acceleration as a function of spectral displacement.

A smooth response spectrum is superimposed on each of the response spectra graphs shown in Figures C.2 through C.19. These smooth response spectra were

constructed using the FEMA 273 guidelines (ATC, 1997) described in Section 4.4 of Chapter 4. The values of the parameters that determine the shape of the smooth response spectra are provided in Tables C.4 through C.6.

EQ Label




Scale Factor

T3 (sec)

to4 (sec)

PGA5 ig)

Was this article helpful?

0 0

Post a comment