Seismic monitoring capability in the Soviet Union using hypothetical regional networks: Final technical report, February 1986-August 1987

The capabilities of hypothetical networks designed for nuclear test monitoring in the Soviet Union are described. This is done with a network simulation program that is an extension of the SNAPD program for computing the detection threshold and location uncertainty for a specified seismic network. Our hypothetical networks all include 13 high quality arrays external to the Soviet Union. To this network we add from 10 to 50 evenly spaced arrays and stations within the Soviet Union and determine the effects on monitoring capability. We have extended the location uncertainty calculations in SNAPD to include the contribution of azimuth estimates from arrays and three-component stations and to estimate the uncertainty from a suite of Monte-Carlo realizations of the sets of detecting stations. The key to valid simulations is accurate normalization to empirical experience; that is, to select appropriate values for the parameters controlling the simulation results. These include the amplitude-distance curves, the station detection thresholds, the source assumptions, and the assumed variance. Our normalization is based om empirical experience with the NORESS array in Norway, but we carefully explore the effect of changing the important assumptions. As a result of this analysis, we are able to place bounds on themore » detection threshold and location uncertainty for these hypothetical networks which are expressed in tables and plots. We find that achieving a detection threshold of m/sub b/ 2.5 throughout the Soviet Union would require one or more of the following: (1) 20 or more in-country arrays of NORESS type; (2) seismic wave propagation in the Soviet Union that is more efficient than we observe near NORESS; (3) station sites that have significantly lower detection thresholds than NORESS; (4) a significantly improved ability to process and analyze seismic data.« less