Surface-Related Multiple Attenuation Based On Wave Equation

In seismic data processing the interference of multiple reflections is still a large problem, especially for marine dataset. Strong multiples can lead to fault seismic interpretation and finally lead to dry-well. So it is very important and necessary for multiple suppression. The research of this dissertation is financed by the Chinese National Natural Science Foundation (49974028).Over the years the SRMA (surface-related multiple attenuation) has proved to be very useful and powerful because all types of surface-related multiples are removed, without the need of a subsurface model, using the data itself as a multiple predictor. A review has been given first for the multiple attenuation techniques. A comprehensive study has been done on the wave equation-based demultiple techniques, especially on SRMA.Starting from the acoustic wave field theory, using the second type of Riley integral, we have obtained wave field extrapolation formula in the frequency-space domain. Regardless of surface reflection, the primary propagates downward, reflects at the subsurface reflector, and then propagates upward. Taking account into the reflection of the free surface, the so-called surface-related multiple has been generated. As the inverse procedure of surface-related multiple generation, surface-related multiple can be removed using the data itself as a multiple predictor. Algorithms, formulas, program flowcharts for surface-related multiple generation and attenuation have been given in detail in this dissertation. From the physical insight, we have explained why surface-related multiple can be predicted and how it is predicted. The surface-related multiple attenuation method can be formulated in an iterative procedure: the output of one iteration step is used as input for the next iteration step. This technique is illustrated with some numerical experiments as well as field data examples. The results show that our SRMA method is very effective and robust. SRMA method seems to be very attractive with the following advantages: completely independent to the velocity, requiring no information about the subsurface, almost automatically running. At last some conclusions has been made and some advice for further research has been given.Based on A.J. Berkhout's pioneering work, this dissertation has made the following innovations: The iterative form for surface-related multiple modeling is presented, which is equivalent to the series form but make the surface-multiple modeling formula more concise and make easier to transfer the algorithm to the practical programming. Fast multi-channel Wiener filter is used for adaptive filtering, which take the horizontal variety into account and has a fast computational speed. Comprehensive and deep discussion for SRMA has been given. Massive work from theory research, algorithm deducing, and program implementing to practical application has been done for this research.