Estimation of the Acoustic-to-Seismic Coupling Ratio Using a Moving Vehicle Source

We present a simple passive technique for estimating the acoustic-to-seismic signal coupling ratio (SAR) in the ground using noise produced by moving vehicles. The seismic signal received on a geophone contains some energy that has propagated as seismic waves and some energy that couples from acoustic waves to seismic waves in the vicinity of the geophone. We use the frequency-domain coherence between the microphone and geophone signals to determine when the seismic signal is predominantly due to acoustic-to-seismic wave coupling. In frequency bands where the microphone and geophone coherence is above 0.8, the ratio of the seismic ground particle velocity to sound pressure-SAR-can be determined with less than 2 dB of error. The method is applied to data from a summer experiment with grass ground cover and at two winter experiments with snow-covered ground. At 100 Hz, the summer analysis yields a SAR value of 1.0 times 10^-5 [(m/s)/Pa]. In addition, at 100 Hz, the two winter tests yield SAR between 0.1 times 10^-5 and 1.0 times 10^-5 [(m/s)/Pa]. In the later winter result, our vehicle-derived SAR estimate is shown to be in excellent agreement with SAR estimates obtained from blank pistol shots. Through the opportunistic exploitation background noise sources, our approach opens the possibility for automatic adaptation of unattended acoustic area, monitoring sensors to changing ground conditions.