Approximate separation of pure-mode and converted waves in 3-C reflection seismics by τ-p transform

Abstract

The use of multicomponent receivers allows one to record the complete elastic wavefield. This is desirable since knowledge of both P- and S-wave characteristics yields better insights into subsurface lithologic and structural rock properties than P-wave knowledge alone. It is often assumed in multicomponent processing that the vertical z-component contains principally pure-mode P-wave arrivals and that the inline horizontal x-component consists mainly of P-SV convertedwave energy. This assumption actually becomes worse with increasing offset. Contaminating energy on either component should ideally be removed to prevent degeneration of stack quality. Often, the x-component is more contaminated than the z-component because P-wave incidence angles (from vertical) are larger than those of P-S waves. This renders processing of the x-component more challenging. Removal of contaminating energy on either component may lead to sharper images. One way of removing such undesired energy is by means of wavefield separation. We propose a simple, approximate wavefield-separation scheme in the τ -p domain to better