ON-SITE SEISMIC YIELD (OSSY) ESTIMATES FOR BRISTOL BASED ON NUCLEAR TO CHEMICAL EXPLOSION SPECTRAL RATIOS AND YIELD SCALING

We have computed a seismic yield estimate for the nuclear explosion BR ISTOL by scaling near-source seismic recordings of BRISTOL with record ings from chemical explosions. Path and near-source coupling effects w ere accounted for by modeling ratios of the spectra of BRISTOL and che mical explosions recorded at identical locations and detonated in near ly the same locations. We estimated BRISTOL's yield, comer frequency, and overshoot by minimizing the difference between observed spectral r atios and model spectral ratios assuming yield scaling. Based on an av erage of spectral ratios at six stations, and assuming an average, gas -filled porosity of 4% in the vicinity of BRISTOL's working point, we obtained a yield of 10 kt with an interstation uncertainty correspondi ng to an increase or decrease by a factor of 1.7. This uncertainty cou ld probably be reduced by using sensors with better low-frequency resp onse and chemical explosions that generated larger signal to noise rat ios (SNR). This yield estimate is in good agreement with a yield estim ate obtained using regional Lg waves of Y = 11. We also find a comer f requency of 2.5 Hz and no evidence of overshoot in the seismic source function. Measurements such as these help improve our understanding of explosion sources and may be important for verification of threshold and low-yield nuclear testing treaties because they are less intrusive and potentially less expensive than existing on-site methods such as Continuous Reflectometry for Radius versus Time Experiment (CORRTEX). They are also applicable to explosions of all sizes.