Beamforming on seismic interface waves with an array of geophones on the shallow sea floor

Abstract

The problem of locating very low frequency sound sources in shallow water is made difficult by the interaction of propagating acoustic waves with the sea floor. It is known that low frequency sound waves enter the bottom and are converted to a variety of compressional and shear wave types, including seismic interface waves. One of the latter is the Scholte wave, which travels in an elliptical orbit along the sediment-water interface. Scholte waves, although dispersive, often have speeds very much slower than the speed of sound in water. Slow wave speeds and the attendant short wavelengths suggest that low frequency beamforming and source localization with sea floor geophones can be accomplished with relatively small arrays when compared with hydrophone arrays in the water column. To test the feasibility of this approach, experiments were carried out in the shallow water of the Malta Channel of the Straits of Sicily where the Scholte wave speed was some 10 to 20 times slower that the speed of sound in water. A linear array of ten vertically gimballed geophones was deployed and measurements were made on propagating seismic wave fields generated by explosive shots. The spatial coherence of the dispersive Scholle waves across the 40-m array was found to be above 0.9 for all shots, while the spatial coherence of the noise fell to 0.5 over a distance of 18 m along the array, indicating good prospects for array beamforming and noise rejection. Frequency dependent group velocities were obtained from the dataset and used to obtain phase velocities needed to implement an algorithm for dispersive beamforming. Since the phase velocities were quite low (130-200 m/s), narrow beams were formed at very low frequencies with this small array; half-power widths of 22° at 7 Hz and 16° at 11 Hz were obtained. The resulting directivities, beam patterns, and sidelobe characteristics are in excellent agreement with array theory, which suggests that coherent processing is a viable technique on which to base new applications for seismic arrays on the sea floor. Supporting material on the geophysics of Scholte waves is also presented, as are calculations of the wave field at the site, made possible by inversion of the velocity dispersion curves.