Statistical Discrimination. Higher Mode Detection Studies

• Main Author: Swindell, William H
• Other Authors: TEXAS INSTRUMENTS INC DALLAS SERVICES GROUP
• Format: Text
• Language: English
• Published: 1970
• Subjects: Seismology; Seismic Detection and Detectors; *SEISMOLOGICAL STATIONS; *UNDERGROUND EXPLOSIONS; ALASKA; ALEUTIAN ISLANDS; CALIFORNIA; CHINA; DETECTION; EARTHQUAKES; ELECTROMAGNETIC WAVE FILTERS; EUROPE; JAPAN; MONTANA; NOISE; NUCLEAR EXPLOSIONS; POWER SPECTRA; RAYLEIGH WAVES; SEISMOMETERS; SIGNAL TO NOISE RATIO; SPECTRUM SIGNATURES; TIME SERIES ANALYSIS; *LARGE APERTURE SEISMIC ARRAY; *LASA(LARGE APERTURE SEISMIC ARRAY); LONG PERIOD SEISMOMETERS; SPECTRUM ANALYSIS; VELA UNIFORM; Alaska; Aleutian Islands
• Online Access: http://www.dtic.mil/docs/citations/AD0880494; http://oai.dtic.mil/oai/oai?&verb=getRecord&metadataPrefix=html&identifier=AD0880494

The report presents results of applying a maximum entropy power spectrum analysis technique to detect higher order mode surface waves. The algorithm used partitioned each seismogram into time gates representing average group velocities, and spectral analysis was used to determine the periods of the data in the gates. The technique was tested successfully on artificial seismograms containing the Rayleigh and three higher modes of surface waves, with and without additive noise. Maximum entropy power spectrum analysis techniques were applied to an ensemble of 29 long-period seismograms obtained from LASA beams. The long-period seismograms were of both continental and continental-oceanic paths. While several instances of possible higher mode waves were revealed by the analysis, no consistency was found in the occurrence of these modes. Major conclusions were that ME spectrum analysis provides finer details in seismogram spectra than are otherwise available and that it provides an effective method for computing group-velocity frequency curves. See also Report no. 2, AD0871116.