Regional Seismic Discrimination Optimization With and Without Nuclear Test Data: Western U.S. Examples

The western U.S. has abundant natural seismicity, historic nuclear explosion data, and widespread mine blasts, making it a good testing ground to study the performance of regional source-type discrimination techniques. We have assembled and measured a large set of these events to systematically explore how to best optimize discrimination performance. Nuclear explosions can be discriminated from a background of earthquakes using regional phase (Pn, Pg, Sn, Lg) amplitude measures such as high frequency P/S ratios. The discrimination performance is improved if the amplitudes can be corrected for source size and path length effects. We show good results are achieved using earthquakes alone to calibrate for these effects with the MDAC technique (Walter and Taylor, 2001). We show significant further improvement is then possible by combining multiple MDAC amplitude ratios using an optimized weighting technique such as Linear Discriminant Analysis (LDA). However this requires data or models for both earthquakes and explosions. In many areas of the world regional distance nuclear explosion data is lacking, but mine blast data is available. Mine explosions are often designed to fracture and/or move rock, giving them different frequency and amplitude behavior than contained chemical shots, which seismically look like nuclear tests. Here we explore discrimination performance differences between explosion types, the possible disparity in the optimization parameters that would be chosen if only chemical explosions were available and the corresponding effect of that disparity on nuclear explosion discrimination. There are a variety of additional techniques in the literature also having the potential to improve regional high frequency P/S discrimination. We explore two of these here: three-component averaging and maximum phase amplitude measures. Typical discrimination studies use only the vertical component measures and for some historic regional nuclear records these are all that are available. However S-waves are often better recorded on the horizontal components and some studies have shown that using a three-component average or a vertical-P/horizontal-S or other three-component measure can improve discrimination over using the vertical alone (e.g. Kim et al. 1997; Bowers et al 2001). Here we compare the performance of vertical and three-component measures on the western U. S. test set. A complication in regional discrimination is the variation in P and S-wave propagation with region. The dominantly observed regional high frequency S-wave can vary with path between Sn and Lg in a spatially complex way. Since the relative lack of high frequency S-waves is the signature of an explosion, failing to account for this could lead to misidentifying an earthquake as an explosion. The regional P phases Pn and Pg vary similarly with path and also with distance, with Pg sometimes being a strong phase at near regional distances but not far regional. One way to try and handle these issues is to correct for all four regional phases but choose the phase with the maximum amplitude. A variation on this strategy is to always use Pn but choose the maximum S phase (e.g. Bottone et al. 2002). Here we compare the discrimination performance of several different (max P)/(max S) measures to vertical, three-component and multivariate measures. Our preliminary results show that multivariate measures perform much better than single ratios, though transportability of the LDA weights between regions is an issue. Also in our preliminary results, we do not find large discrimination performance improvements with three-component averages and maximum phase amplitude measures compared to using the vertical component alone