Regional magnitude scaling, transportability, and M-s : m(b) discrimination at small magnitudes

Data sets of m(b)(Pn) and m(b)(Lg) measurements are presented for three con tinental regions in order to investigate scaling relationships with moment magnitude M-w and event discrimination at small magnitudes. Compilations of published measurements are provided for eastern North American and central Asian earthquakes, and new measurements are reported for earthquakes locat ed in western United States. Statistical tests on M-w:m(b) relationships sh ow that the m(b)(Lg) scale of NUTTLI (1973) is transportable between tecton ic regions, and a single, unified M-w:m(b)(Lg) relationship satisfies obser vations for M-w similar to4.2-6.5 in all regions. A unified relationship is also developed for nuclear explosions detonated at the Nevada Test Site an d test sites of the former Soviet Union. Regional m(b) for explosions scale at higher rates than for earthquakes, and of significance is the finding t hat m(b)(Pn) for explosions scales at a higher rate than m(b)(Lg). A model is proposed where differences in scaling rates are related to effects of sp ectral overshoot and near-field Rg scattering on the generation of Pn and L g waves by explosions. For earthquakes, m(b)(Pn) and m(b)(Lg) scale similar ly, showing rates near 1.0 or 2/3 . log(10)M(o) (seismic moment). M-w:m(b)(Lg) scaling results are converted to unified M-s:m(b)(Lg) relation ships using scaling laws between log M-o and M-s. For earthquakes with M-s greater than 3.0, the scaling rate is 0.69 . M-s, which is the same as it i s for nuclear explosions if M-s is proportional to 1.12 . log M-o as determ ined by NTS observations. Thus, earthquake and explosion populations are pa rallel and separated by 0.68 m(b) units for large events. For small events (M-s < 3.0), populations may converge or diverge depending on the tectonic region in which earthquakes occur and the scaling rate of explosions at sma ll yields. Earthquakes scale as 0.64 and 0.75 on M-s:m(b)(Lg) plots for sta ble and tectonic regions, respectively. While the scaling rate for explosio ns is similar to0.69, this value is uncertain due to paucity of M-o observa tions at small yields. Measurements of [m(b)(P) - m(b)(Lg)] for earthquakes in the western United States have an average value of -0.33 +/- .03 m(b) u nits, in good agreement with Nuttli's estimate of m(b) bias for NTS. This r esult suggests that Nuttli's method for estimating test site bias can be ex tended to earthquakes to make estimates of bias on regional scales. In addi tion, a new approach for quick assessments of regional bias is proposed whe re M-s:m(b)(P) observations are compared with M-s:m(b)(Lg) relationships. C atalog M-s:m(b)(P) data suggest that m(b) bias is significant for tectonic regions of southern Asia, averaging about -0.4 m(b) units.