Mrg: A Magnitude Scale for 1 s Rayleigh Waves at Local Distances with Focus on Yield Estimation
Technical Report,28 Sep 2009,30 Jun 2016
WESTON GEOPHYSICAL CORP LEXINGTON MA LEXINGTON United States
Pagination or Media Count:
The accurate estimation of yields from small chemical and nuclear explosions represents a difficult and continuing problem for the Nuclear Explosion Monitoring NEM community. P-wave spectra, Pn amplitudes, Lg amplitudes, intermediate-period surface wave amplitudes, or coda techniques are some of the methods that have been proposed for estimating the yields of small explosions recorded at regional distances. However, the uncertainty on seismic yield estimates can be large, and reducing the uncertainty may require a priori information about source media, knowledge of the emplacement depth, and calibrations for path and site effects. There is no general consensus as to which method for seismic yield estimation works best for all nuclear test sites. For monitoring at local and near-regional distances, we propose a methodology for yield estimation based on magnitudes of short-period, fundamental mode Rayleigh waves Rg. At local distances, Rg can be the largest amplitude seismic arrival observed from shallow explosions, mining explosions, and shallow earthquakes. Adushkin 2001 demonstrated the ability to use Rg amplitudes at near-regional distances to provide accurate seismically-estimated yields. He corrected Rg amplitudes from Semipalatinsk underground explosions for attenuation, geometric spreading, and station-specific effects. Using the corrected Rg amplitudes, he estimated explosive yield within 20 of the true yields for the explosions. In this report, we develop a new formula for short-period roughly 1 s surface wave magnitudes, called MRg, using the methods of Russell 2006. Attenuation coefficients for short-period surface waves have been derived from diverse tectonic settings to calibrate the new MRg formula. We demonstrate the utility of the formula at estimating magnitudes for small explosions, and then correlate the magnitudes to yield with uncertainty estimates.
- Seismic Detection and Detectors
- Operations Research