5: Multicomponent Seismic Processing
Computer processing of multicomponent seismic data is substantially more complex than conventional P-wave seismic data processing. The increase in processing complexity is greater than the nine-fold increase in data volume would suggest. This arises because of the need to process the different components in a mutually compatible fashion since much of the interpretative value of multicomponent data derives from the comparison of reflection arrival times and amplitudes on different components. For example, a lateral change in reflection strength of an event on a P-P wave component that correlates with a SH-SH wave event whose reflection amplitude is not changing could signal a change in pore fluid type within a reservoir IF one is certain that the amplitude variations are indicative of changes in geology, and not the result of processing the seismic data using erroneous analysis parameters or procedures. Thus, the goal of multicomponent data processing can be stated as follows:
The similarities and differences in reflection amplitudes, phases, and arrival times in the processed P- and S-wave sections should be indicative of the similarities and differences in the propagation of P- and S-waves in the earth, and not simply artifacts of data processing.
Figures & Tables
Many prospective basins of the world are, or will soon become, “mature” in the exploration sense. Increasingly we must resort to nonconventional technology and techniques to uncover the remaining hydrocarbon reserves that are often found in complex or subtle traps. Multicomponent seismology-the use of concurrent, combined shear (S)- and compressional (P)-wave seismology-is gaining acceptance in the exploration community as one tool that can provide direct measurements of subsurface rock properties. These measurements can detect new hydrocarbon accumulations, and aid in the efficient and economic development of newly found or existing reservoirs by providing detailed maps of reservoir porosity, lithology, and pore fluid distributions.