Showing papers by "David M. Boore published in 1983"

Stochastic simulation of high-frequency ground motions based on seismological models of the radiated spectra

Abstract: Theoretical predictions of seismic motions as a function of source strength are often expressed as frequency-domain scaling models. The observations of interest to strong-motion seismology, however, are usually in the time domain (e.g., various peak motions, including magnitude). The method of simulation presented here makes use of both domains; its essence is to filter a suite of windowed, stochastic time series so that the amplitude spectra are equal, on the average, to the specified spectra. Because of its success in predicting peak and rms accelerations (Hanks and McGuire, 1981), an ω -squared spectrum with a high-frequency cutoff ( f m), in addition to the usual whole-path anelastic attenuation, and with a constant stress parameter (Δ σ ) has been used in the applications of the simulation method. With these assumptions, the model is particularly simple: the scaling with source size depends on only one parameter—seismic moment or, equivalently, moment magnitude. Besides peak acceleration, the model gives a good fit to a number of ground motion amplitude measures derived from previous analyses of hundreds of recordings from earthquakes in western North America, ranging from a moment magnitude of 5.0 to 7.7. These measures of ground motion include peak velocity, Wood-Anderson instrument response, and response spectra. The model also fits peak velocities and peak accelerations for South African earthquakes with moment magnitudes of 0.4 to 2.4 (with f m = 400 Hz and Δ σ = 50 bars, compared to f m = 15 Hz and Δ σ = 100 bars for the western North America data). Remarkably, the model seems to fit all essential aspects of high-frequency ground motions for earthquakes over a very large magnitude range . Although the simulation method is useful for applications requiring one or more time series, a simpler, less costly method based on various formulas from random vibration theory will often suffice for applications requiring only peak motions. Hanks and McGuire (1981) used such an approach in their prediction of peak acceleration. This paper contains a generalization of their approach; the formulas used depend on the moments (in the statistical sense) of the squared amplitude spectra, and therefore can be applied to any time series having a stochastic character, including ground acceleration, velocity, and the oscillator outputs on which response spectra and magnitude are based .

Variability in ground motions: root mean square acceleration and peak acceleration for the 1971 san fernando, california, earthquake

Abstract: Data from the 1971 San Fernando, California, earthquake provided the opportunity to study the variation of ground motions on a local scale. The uncertainty in ground motion was analyzed by studying the residuals about a regression with distance and by utilizing the network of strong-motion instruments in three local geographic regions in the Los Angeles area. Our objectives were to compare the uncertainty in the peak ground acceleration (PGA) and root mean square acceleration (RMSa) about regressions on distance, and to isolate components of the variance. We find that the RMSa has only a slightly lower logarithmic standard deviation than the PGA and conclude that the RMSa does not provide a more stable measure of ground motion than does the PGA (as is commonly assumed). By conducting an analysis of the residuals, we have estimated contributions to the scatter in high-frequency ground motion due to phenomena local to the recording station, building effects defined by the depth of instrument embedment, and propagation-path effects. We observe a systematic decrease in both PGA and RMSa with increasing embedment depth. After removing this effect, we still find a significant variation (a standard deviation equivalent to a factor of up to 1.3) in the ground motions within small regions (circles of 0.5 km radius). We conclude that detailed studies which account for local site effects, including building effects, could reduce the uncertainty in ground motion predictions (as much as a factor of 1.3) attributable to these components. However, an irreducible component of the scatter in attenuation remains due to the randomness of stress release along faults during earthquakes. In a recent paper, Joyner and Boore (1981) estimate that the standard deviation associated with intra-earthquake variability corresponds to a factor of 1.35.

Comments on "new attenuation relations for peak and expected accelerations of strong ground motion," by b. a. bolt and n. a. abrahamson

Abstract: Bolt and Abrahamson (1982) imply that we neglected to provide sufficient flexibility in the form of the ground-motion attenuation relationship used in our recent study (Joyner and Boore, 1981). We disagree with this implication, and we want to be sure that there is no misunderstanding concerning our views in the matter. We deliberately constrained the geometric spreading coefficient to 1 (i.e., 1/R spreading) in our relationship because we did not believe that the data set permitted physically meaningful, simultaneous determination of a spreading coefficient and a coefficient of anelastic attenuation. [We believe that the relationship used by Bolt and Abrahamson has too many free parameters; this is evidenced by the physically implausible values they obtain for the geometric spreading coefficient, which ranges from -0.2 to +0.38 (corresponding to values of 0.1 to -0.19 for the parameter c in their equation 9).] They contend that because our functional form is insufficiently flexible, our near-source estimates of motion are largely determined by the data at larger distances. In an attempt to prove the contention, they repeat our analysis removing the data points from stations at distances less than 8 km and report parameter values very close to what we obtained for the whole data set. This, however, only proves that the truncated data set and the whole data set are compatible with the same set of parameters, nothing more. The best way to determine how well our curves fit the data at short distance is to examine the plots of residuals given in our paper. A detailed discussion of our method and results is given in that and a subsequent paper (Boore and Joyner, 1982). Contrary to their contention, our near-source estimates for magnitudes in the 5.0 to 7.0 range are not largely determined by the data at larger distances. Indeed, for magnitudes 5.5 and 6.5, our near-source (0.1 km) estimates agree within 18 and 2 per cent, respectively, with the estimates they obtained for the magnitude ranges 5.0 to 5.9 and 6.0 to 6.9 using their more flexible functional form. Above magnitude 7.0, our near-source estimates are determined by the data points at large distance and a functional relationship with distance that is controlled by parameters determined by fitting the whole data set. Thus, a combination of the distant data points at magnitudes greater than 7.0 with the near-source data points at magnitudes less than 7.0 determines our near-source estimates for magnitudes greater than 7.0. If one does not wish to use such a strategy, one must forego near-source estimates for large magnitudes. As Bolt and Abrahamson demonstrate, an attempt to make such estimates using only data for magnitudes greater than 7.0 leads to the unlikely result that the estimate for magnitude 7.0 to 7.7 is less than that for 5.0 to 6.0. A second point requiring comment is their statement (p. 2314) that "available acceleration data do not imply a systematic increase in peak acceleration with magnitude in the near-source region . . ." We believe that this statement is contradicted by their own results, which they appear to have misinterpreted by virture of confusing the standard deviation of an individual observation with the standard error of the mean. They obtain intercepts at x = 0.1 km of 0.34 g for the magnitude range 5.0 to 5.9 and 0.52 g for the range 6.0 to 6.9. They then take the data points for x < 10 km in the two magnitude ranges, compute standard deviations