While the interpretation of reflected P-waves on seismic data remains the main vehicle for seismic interpretation, there are other signals in seismic reflection recordings that can be fully utilized in seismic inversion. There are reflection signals that are due to the conversion of P-wave energy to S-wave energy in transmission followed by conversion from S-wave to P-wave upon reflection. These waves, known as PSSP waves, have significant amplitude and normal moveout (NMO), and are seen on reflection records at wide offset. We model PSSP waves by ray tracing and finite-difference wave equation computations. While PSSP amplitudes are essentially zero at normal incidence for flat reflectors, their energy is considerable at larger offsets. Also, the PSSP energy for non-flat reflectors will generally be nonzero at zero offset. In addition to identification of the PSSP modes, there is the challenge of utilizing this energy for estimation of seismic velocities. While the NMO for PSSP arrivals allows it to be suppressed through stacking in imaging P-wave reflections, it is feasible that full waveform inversion could be implemented for utilizing the PSSP energy as useful signal rather than treating it as undesirable "noise".
View full article as PDF (0.63 Mb)