Enhancement and modelling of signals from natural piezoelectric targets

Electromagnetic signals from piezoelectric targets in the earth can be generated using seismic sources and measured with electric or magnetic field receivers. The signals are typically small compared to the ambient electromagnetic noise and are difficult to identify in unprocessed records. Three data processing algorithms involving stacking, low-pass filtering, and sinusoid subtraction have been developed to enhance the signal-to-noise ratio of piezoelectric data acquired during field experiments. In addition, an analytic modelling technique has been developed to investigate the relationship between seismic and piezoelectric signals. The stacking technique, designed for use with repetitive seismic sources, employs a robust triggering algorithm that enables it to be used effectively even when the trigger signal is poor. High frequency noise is attenuated using a zero phase frequency domain low-pass filter with variable cut-off frequency and slope. The sinusoid subtraction technique is used to remove powerline noise which occurs at frequencies of 60 Hz and its harmonics. The amplitude and phase of each harmonic are estimated by calculating the Fourier series coefficients for that frequency. A sinusoid having the estimated amplitude and phase is then subtracted from the data to remove the harmonic. Remarkable improvements in the signal-to-noise ratio have been achieved by sinusoid subtraction as powerline noise levels typically exceed piezoelectric signal amplitudes by factors ranging from five to a few hundred times. For purposes of analytic modelling, a piezoelectric target is represented by a number of spheres each of which become independently polarized during the passage of a spherical elastic wave. The electric potential at a point in a uniform conductive medium surrounding the target is estimated by summing the potentials due to each of the polarized spheres. It is shown that the form of the electric potential time series generated by a uniformly polarized target is approximately proportional to the first derivative of the elastic wave particle velocity. In contrast, the potentials generated by non-uniformly polarized targets, representing quartz veins, are approximately proportional to the particle velocity itself. A distinct pulse of electric potential is also generated when the elastic wave reaches the outer limits of a vein-like target; such signals could yield important information on target dimensions if they could be identified in real piezoelectric data.Science, Faculty ofEarth, Ocean and Atmospheric Sciences, Department ofGraduat