3-D Elastic Time-Reverse Imaging of a Linear Void Anomaly

Madeleine Pels , Jeffrey Shragge, Aaron Girard

Imaging meter-scale subsurface heterogeneities such as void spaces remains a difficult task for most established near-surface seismic methods. One strategy for addressing this challenge is to isolate and use surface waves backscattered from lateral heterogeneities to identify and characterize the sources of scattered energy. Seismic methods associated with this strategy, though, often require dense geophone sampling, demand significant user-intensive preprocessing, and/or involve large computation run times. These limitations motivate the development of more effective and efficient near-surface seismic analysis approaches. One way to address such challenges is to use elastic time reverse imaging (E-TRI), a migration method originally proposed for microseismic event location. E-TRI propagates two wavefields in reverse time and then applies an imaging condition to generate an image of the source locations of scattered energy. When adapted for detecting near-surface heterogeneities, the two wavefields are the estimated outward-propagating transmission and inward-propagating scattered surface-wave energy. These wavefields are used in an elastic imaging condition that correlates temporally and spatially collocated energy to image the sources of surface-wave scattering. Synthetic tests demonstrate the ability of E-TRI to image meter-scale heterogeneities at seismic wavelengths between 2.0-5.0$\times$ the characteristic scale length of the introduced subsurface anomalies. Results from a field test using 3-D seismic data acquired at the Yuma Proving Grounds highlight E-TRI's ability to detect a purpose-built long linear void space of meter-scale cross-sectional area located at 10~m depth. The results suggest that E-TRI, perhaps with improved depth sensitivity through more advanced imaging conditions, would be a reliable near-surface seismic method for characterizing small-scale heterogeneity in complex geological environments.