Regional surface wave magnitude and moment determination methods applied to nuclear explosions at the Nevada test site : implications for yield estimation and seismic discrimination

Citation

Woods, Bradley Brett (1994) Regional surface wave magnitude and moment determination methods applied to nuclear explosions at the Nevada test site : implications for yield estimation and seismic discrimination. Dissertation (Ph.D.), California Institute of Technology. doi:10.7907/P212-Q044. https://resolver.caltech.edu/CaltechETD:etd-04212006-165925

Abstract

NOTE: Text or symbols not renderable in plain ASCII are indicated by [...]. Abstract is included in .pdf document. This thesis examines the use of regional surface-wave data to measure the long-period source spectrum of underground nuclear explosions for the purposes of yield determination and seismic discrimination. It is demonstrated that regional (D < 2500 km) fundamental-mode Rayleigh and Love waveforms can be modeled with considerable accuracy. The procedure for modeling regional earth structure for such seismograms by inverting surface-wave dispersion data is described. This technique is a hybrid of preexisting surface-wave analysis and inversion methods. Theoretical path corrections are determined from the Green's function for a given modeled path. A method is described to obtain consistent, stable, time-domain surface-wave magnitude ([...]) or seismic moment ([...]) measurements from poorly dispersed regional Rayleigh waves. Source parameters for 190 Nevada Test Site explosions are determined using these methods. Observations demonstrate that the measurement/detection threshold for regional surface-waves is [...] > 4.0 (Yield = 1 kt)--a significant improvement over classical teleseismic [...] measurements. The results indicate that the [...] (or log [...]) - yield scaling relationship is near unity and constant for explosions of all measurable sizes. Site effects are also investigated to determine the portability of such surfacewave measurements. Spectral-domain moment estimates also were performed on the digital portion of the data set. Besides obtaining an average scalar moment from Rayleigh wave amplitudes, the isotropic (explosive source) and deviatoric moment (double-couple source generated by tectonic release) components were determined by a joint inversion of Rayleigh and Love wave amplitude and phase data. Although in the most general case the inversion solution is non-unique, constraining the depth of the deviatoric source to be equal to that of the explosion and assuming a vertical strike-slip orientation yields a unique linear inversion solution. The spectral moment estimates are similar to the time-domain values, although the spectral-domain moment variances are appreciably smaller than the time-domain ones. A regional short-period vs. long-period seismic discriminant is developed using the ratio of the seismic moment to local magnitude ([...]). This discriminant successfully separates the explosion and earthquake populations at all measurable source sizes, so that for a given seismic moment source level, an explosion has an [...] 0.5 magnitude units larger than a comparable-sized earthquake.

Thesis Files