A study was made to determine if significant phase differences exist between Rayleigh waves which have traveled almost identical paths from the USSR Shagan River test site. Surface waves from five explosions recorded at six SRO/ASRO digital stations were used in the analysis. The explosion of 4 August 1979 was selected as a reference, and at each site the phase spectra of the Rayleigh waves from the other five events were compared to that of the reference. The technique of phase-matched filtering was used to analyze the signals. This technique reduces the effects of multipathing and removes phase differences due to dispersion along slightly different travel paths.
Each epicenter had been relocated by using calibration data from the cratering shot of 15 January 1965, and remaining errors in location and origin time are considered to be extremely small. Seismograms were analyzed for surface waves recorded at: Matsushiro, Japan (MAJO); Shillong, India (SHIO); Kabul, Afghanistan (KAAO); Ankara, Turkey (ANTO); Grafenburg, West Germany (GRFO); and Albuquerque, New Mexico (ANMO).
Results of the study indicate that Rayleigh waves from some Shagan River explosions have undergone large phase shifts relative to Rayleigh waves which have traveled almost identical paths from other Shagan River explosions. In some cases, the phase shifts can be interpreted as complete phase reversals with associated time delay. In particular, as compared to the explosion of 4 August 1979, Rayleigh waves from the explosion of 7 July 1979 are reversed in polarity at KAAO and ANTO and are reversed in polarity and delayed at GRFO, SHIO, and MAJO.