Observational data on the attenuation of short‐period Rayleigh waves in North America east of the Rocky Mountains yield the following average values for the coefficient of anelastic attenuation: γ = 0.07 deg−1 for 1‐sec‐period waves and γ = 0.10 deg−1 for waves with a maximum particle velocity in the period range 3–12 sec. By way of comparison, the amplitude data that form the basis of Richter's empirical local magnitude scale for southern California give γ = 0.60 deg−1. Differences in γ values are sufficient to explain the observation that earthquakes in the eastern United States have a radius of perceptibility as much as 10 times larger than that of earthquakes of the same magnitude in the western United States. Theoretical curves of log A/T versus log Δ are not linear. Thus magnitude formulas of the type M = B + C (log Δ) + log A/T are valid only over a limited range of distance, for which the curve can be approximated by a straight line. Formulas of this kind, which give mb and Ms from short‐period Rayleigh waves, are proposed for eastern North America.

A methodology is presented for determining the yield of underground nuclear explosions from Lg wave amplitudes. The methodology is applied to Nevada Test Site (NTS) explosions, for which the data from short‐period, vertical component analog seismographs at three stations are used to develop calibration curves for unsaturated material and water‐saturated rock source conditions. The latter curves are found to provide reasonably accurate estimates of the yields of explosions in other areas of the United States and in the French Sahara, suggesting that they may be applicable to all continental areas. If so, they also can provide an estimate of the bias of mb(P) magnitudes between different continental sites. For example, the Lg data from NTS explosions indicate a 0.31±0.02 magnitude unit bias between NTS and eastern North America, similar to the approximately 0.33 unit bias found between western and eastern North America previously by use of earthquake data.

No abstract

Based on previous work, a stochastic convolution model is proposed for the Lg coda observed at large epicentral distances (A > 200 km). Coda at these distances differ from local coda in several respects. Effects of various physical processes, such as dispersion, scattering, and mode conversion on coda at large distances are discussed in detail. A spectral ratio method is developed for inversion of Lg coda Q, in which the large variance associated with inversion is greatly reduced. The method is also extended to the inversion of seismic source parameters by jointly using Lg and its coda. We have tested the model and method exhaustively using data from two GDSN stations and found the results to be superior to other methods when those methods used only a small number of stations.

Spectra of more than three hundred short-period, vertical component L, waves of 78 earthquakes in the central United States were determined. From these an empirical relationship between the long-period level of the L, spectrum and the seismic moment was developed. The shape of the reduced spectrum shares some characteristics of Aki's revised model A.There is a uniform relationship between the corner period and the seismic moment, which implies a uniformity of earthquake processes in the central United States over a wide range of event sizes, as opposed to California where wide variations in the corner frequency are observed for source spectra having the same seismic moment. Using the reduced spectra, a method for relating magnitude observations at periods other than 1 s to mb is outlined. 4 > J :: *< a 7 ,I c cd +-I n comer between the wo and w -~ asymptotic trends of the spectra. (b) Event 6, 1962 July 23, showing the To, comer between the wo and w -l trends and the Ti2 comer between the UJ-' and w -~ trends.