Accession Number : ADA305487
Title : Statistical Methodology and Assessment of Seismic Event Characterization Capability.
Descriptive Note : Final rept. 2 Jun 93-2 Sep 95,
Corporate Author : MISSION RESEARCH CORP SANTA BARBARA CA
Personal Author(s) : Fisk, Mark D. ; Gray, Henry L. ; McCartor, Gary D.
PDF Url : ADA305487
Report Date : 31 OCT 1995
Pagination or Media Count : 92
Abstract : This project has focused on developing and applying statistical methods to perform seismic event characterization/identification and on quantifying capabilities with regard to monitoring of a Comprehensive Test Ban. The report is divided into four separate but related sections. First, an automated procedure is described to categorize seismic events, based on multivariate analysis of features derived from seismic waveforms. The procedure is intended to aid CMR subscribers in focussing upon events of interest. Second, preliminary event identification results are presented for a seismic event which occurred on 5 January 1995 in the Southern Ural Mountains region. This was a shallow event which occurred in an area of low seismicity and near where several PNEs had been conducted. Third, various statistics are compiled regarding 1786 seismic events which occurred between 11 January 1995 and 12 February 1995 and were detected by a set of 30 GSETT-3 Alpha stations. The objective of this study is to determine how many events can be screened with high confidence as due to natural seismicity, based on teleseismic measures of depth, Ms:mb and location (offshore/onshore). Of the remaining events, the number with useful regional discriminants is quantified. Last, we evaluate the number of remaining ambiguous events lacking adequate event characterization data to identify them, based on the Alpha network which existed at the time of this study. Fourth, a fundamental problem is addressed of how to utilize multivariate discriminant data from a multistation network in order to optimize the power of the outlier test for fixed false alarm rate. Analytic expressions for the power of three outlier tests are derived and a parametric study is presented in order to determine conditions for which a particular test is favored over the others.
Descriptors : *SEISMIC DETECTION, *NUCLEAR EXPLOSION DETECTION, PARAMETRIC ANALYSIS, MULTIVARIATE ANALYSIS, WAVEFORMS, FALSE ALARMS, SEISMIC WAVES, STATISTICAL PROCESSES, CONFIDENCE LEVEL, SEISMOLOGY, DISCRIMINATE ANALYSIS, ANALYTIC FUNCTIONS.
Subject Categories : Nuclear Warfare
Seismic Detection and Detectors
Distribution Statement : APPROVED FOR PUBLIC RELEASE