Seismic Source Tracking Using Direct 6C Observations and Polarization-Based Wave-Type Fingerprinting

Abstract Determining the back azimuth (BAz) of seismic sources accurately is a critical challenge in seismology, traditionally addressed through methods like P-wave polarization or arrival-time analysis using seismometer networks. However, these approaches often lack precision, particularly in complex wavefields in which scattering effects distort wavefronts, leading to inaccurate localization. To address these challenges, our study introduces a novel solution using six-component (6C) ground-motion measurements. This method determines the BAz of a migrating source using a single-point measurement. We monitored a vibroseis truck’s migration through 480 recorded sweep signals, each lasting 15 s and spanning frequencies from 7 to 120 Hz, across 160 distinct locations. BAz values were derived from the 6C data for each sweep. Because rotational sensors are sensitive primarily to S waves, precision declined as the truck moved farther away due to scattering effects altering the wavefield. To overcome this, we implemented a wave-type fingerprinting algorithm to isolate the SV-type wavefield, significantly improving BAz accuracy. Our results show that this approach reduces deviations from theoretical values to a maximum of 2.2°, compared to deviations of up to 48.6° when the entire wavefield is used. This innovative technique represents a major advancement in seismic source localization, providing improved accuracy even under challenging conditions with complex wavefields.